St Sauver, Jennifer

With increasing number of people living with dementia, the problem of late diagnosis significantly impacts a person’s quality of life while early signs of dementia may provide useful insights to facilitate better treatment plans. With time, this progressive neurodegenerative syndrome could progress from mild cognitive impairment to dementia. A pattern of health conditions can be characterized in unsupervised manner to help predict this progress. As a significant extension to our previous work with streaming clustering model, we consider additional information for predicting dementia onset. With empirical observations, we discover the importance of examining sex and age to predict dementia onset. To this end, we propose a sex-specific model with age-constraint for predicting dementia onset and validate the effectiveness of our models using data from Mayo Clinic Study of Aging (MCSA). The proposed sex-specific models for older adult populations (>=65 years of age) outperformed the previous models with F-score of 77% and 78% for male-specific and female-specific models, respectively. Our experiments of sex-specific temporal clustering of features in older adults demonstrate the potential of more personalized models for early alerts of dementia.

Background: Multiple algorithms with variable performance have been developed to identify dementia using combinations of billing codes and medication data that are widely available from electronic health records (EHR). If the characteristics of misclassified patients are clearly identified, modifying existing algorithms to improve performance may be possible. Objective: To examine the performance of a code-based algorithm to identify dementia cases in the population-based Mayo Clinic Study of Aging (MCSA) where dementia diagnosis (i.e., reference standard) is actively assessed through routine follow-up and describe the characteristics of persons incorrectly categorized. Methods: There were 5,316 participants (age at baseline (mean (SD)): 73.3 (9.68) years; 50.7% male) without dementia at baseline and available EHR data. ICD-9/10 codes and prescription medications for dementia were extracted between baseline and one year after an MCSA dementia diagnosis or last follow-up. Fisher’s exact or Kruskal-Wallis tests were used to compare characteristics between groups. Results: Algorithm sensitivity and specificity were 0.70 (95% CI: 0.67, 0.74) and 0.95 (95% CI: 0.95, 0.96). False positives (i.e., participants falsely diagnosed with dementia by the algorithm) were older, with higher Charlson comorbidity index, more likely to have mild cognitive impairment (MCI), and longer follow-up (versus true negatives). False negatives (versus true positives) were older, more likely to have MCI, or have more functional limitations. Conclusions: We observed a moderate-high performance of the code-based diagnosis method against the population-based MCSA reference standard dementia diagnosis. Older participants and those with MCI at baseline were more likely to be misclassified.

Abstract Background In this longitudinal cohort study, we examined the socio-demographic and psychological predictors of alcohol use initiation during the COVID-19 pandemic in a sample of never alcohol users aged ≥21 prior to COVID-19. Methods Our study population consisted of 56 930 patients aged ≥21, as of 30 March 2019 were collected from a pre-COVID period of 1 year before 31 March 2020, and during-COVID, a period between 1 April 2020 and 30 March 2021. Univariable and multivariable logistic regression models were utilized to examine the roles of socio-demographic variables (gender, age, education, Area Deprivation Index and rural residence) changes in anxiety and depression severity as predictors of alcohol use initiation. Results Age, gender, race, ethnicity, education and rural status were significant predictors in multivariable analysis. A subgroup analysis showed neither anxiety nor depression had a significant association with alcohol use initiation. Conclusion Women, younger individuals, those living in a rural area and people who smoke cigarettes were more likely to initiate alcohol use during the pandemic. Our study has public health and clinical implications such as the need for targeted alcohol use screening and intervention for vulnerable individuals.

Dementia is one of the most prevalent health problems in the aging population. Despite the significant number of people affected, dementia diagnoses are often significantly delayed, missing opportunities to maximize life quality. Early identification of older adults at high risk for dementia may help to maximize current quality of life and to improve planning for future health needs in dementia patients. However, most existing risk prediction models predominantly use static variables, not considering temporal patterns of health status. This study used an attention-based time-aware model to predict incident dementia that incorporated longitudinal temporal health conditions. The predictive performance of the time-aware model was compared with three traditional models using static variables and demonstrated higher predictive power.

Reproducibility is an important quality criterion for the secondary use of electronic health records (EHRs). However, multiple barriers to reproducibility are embedded in the heterogeneous EHR environment. These barriers include complex processes for collecting and organizing EHR data and dynamic multi-level interactions occurring during information use (e.g., inter-personal, inter-system, and cross-institutional). To ensure reproducible use of EHRs, we investigated four information quality dimensions and examine the implications for reproducibility based on a real-world EHR study. Four types of IQ measurements suggested that barriers to reproducibility occurred for all stages of secondary use of EHR data. We discussed our recommendations and emphasized the importance of promoting transparent, high-throughput, and accessible data infrastructures and implementation best practices (e.g., data quality assessment, reporting standard).

Abstract Aims The objective of this study is to longitudinally assess sociodemographic and psychological correlates of increased alcohol use during the coronavirus disease of 2019 (COVID-19) period among adolescents and young adults. Methods Pre-COVID period is defined as the 1-year period on or before 31 March 2020, and during-COVID period is defined as the period from 1st April 2020 to 30 March 2021. Univariable logistic regression models are used to evaluate the association of demographic characteristics, Area Deprivation Index (ADI), rurality, changes in Patient Health Questionnaire (PHQ-9) and Generalized Anxiety Disorder (GAD-7) scale severity, and the risk of increased alcohol consumption (binge drinking, number of drinks and drinking regularity) from pre-COVID to during-COVID period. Results and conclusion Our study found that worsened anxiety symptoms, older age, being in college and current cigarette smoking status were associated with increased alcohol use among youth during the pandemic year. Socioeconomic position (measured by ADI) and rural status were not found to be associated with increased alcohol use among adolescents and young adults.

Abstract Background Cognitive function is essential to effective self‐management of heart failure (HF). Alzheimer's disease and Alzheimer's disease‐related dementias (AD/ADRD) can coexist with HF, but its exact prevalence and impact on health care utilization and death are not well defined. Methods Residents from 7 southeast Minnesota counties with a first‐ever diagnosis code for HF between January 1, 2013 and December 31, 2018 were identified. Clinically diagnosed AD/ADRD was ascertained using the Centers for Medicare and Medicaid (CMS) Chronic Conditions Data Warehouse algorithm. Patients were followed through March 31, 2020. Cox and Andersen‐Gill models were used to examine associations between AD/ADRD (before and after HF) and death and hospitalizations, respectively. Results Among 6336 patients with HF (mean age [SD] 75 years [14], 48% female), 644 (10%) carried a diagnosis of AD/ADRD at index HF diagnosis. The 3‐year cumulative incidence of AD/ADRD after HF diagnosis was 17%. During follow‐up (mean [SD] 3.2 [1.9] years), 2618 deaths and 15,475 hospitalizations occurred. After adjustment, patients with AD/ADRD before HF had nearly a 2.7 times increased risk of death, but no increased risk of hospitalization compared to those without AD/ADRD. When AD/ADRD was diagnosed after the index HF date, patients experienced a 3.7 times increased risk of death and a 73% increased risk of hospitalization compared to those who remain free of AD/ADRD. Conclusions In a large, community cohort of patients with incident HF, the burden of AD/ADRD is quite high as more than one‐fourth of patients with HF received a diagnosis of AD/ADRD either before or after HF diagnosis. AD/ADRD markedly increases the risk of adverse outcomes in HF underscoring the need for future studies focused on holistic approaches to improve outcomes.