Citations

Objective To identify the optimal incentive protocol for maximising participation while managing study costs during the Voyage trial. Design Prospective cohort (Voyage trial) of colorectal cancer (CRC) incidence and mortality outcomes in individuals screened with multitarget stool DNA (mt-sDNA) served as the population. A subset was randomised to receive postage stamps as a pre-consent incentive, or as a post-consent incentive after completion of the consent and questionnaire. Descriptive statistics from year 1 are reported. Results During year 1 of the Voyage trial, a total of 600 258 individuals with mt-sDNA orders received at Exact Sciences Laboratories were randomly selected and invited to participate. Of those, 26 429 (4.4%) opted in, 14 365 of whom (54.3%) consented. The opt-in and consent samples were similar to the target population with respect to sex but differed by geographic residence and age (p<0.001). For the embedded incentive experiment, 2333 were randomised to the pre-incentive arm, while 2342 were randomised to the post-incentive arm. Overall consent rate in the incentive trial was 56.4% (60.9% for the pre-consent incentive arm (1421/2333) vs 52.0% for the post-consent incentive arm (1217/2342), p<0.001). Cost reduction was observed for the pre-consent incentive group, and higher response rates were seen among older versus younger individuals. Conclusions Pre-consent incentive option was associated with a higher participation rate and lower costs and was used for the remainder of study recruitment. CRC incidence and mortality vary with age; thus, adjusting for differential participation by age and region will be important in analyses of Voyage data. Trial registration number NCT04124406.

Objectives Primary human papillomavirus (HPV) testing by clinician-collection is endorsed by U.S. guideline organizations for cervical cancer screening, but uptake remains low and insights into patients’ understanding are limited. This study aims to primarily address patient awareness of primary HPV screening by clinician-collection and acceptance of primary HPV screening by clinician- and self-collection, and secondarily assess factors associated with awareness and acceptance. Setting Primary care practices affiliated with an academic medical center. Methods A cross-sectional survey study of screening-eligible women aged 30–65 years was conducted to assess awareness and acceptability of primary HPV screening. We analyzed bivariate associations of respondent characteristics with awareness of primary HPV screening by clinician-collection, willingness to have clinician- or self-collected primary HPV testing, and reasons for self-collection preference. Results Respondents (n = 351; response rate = 23.4%) reported cervical cancer screening adherence of 82.8% but awareness of clinician-collected primary HPV as an option was low (18.9%) and only associated with HPV testing with recent screening ( p = 0.003). After reviewing a description of primary HPV screening, willingness for clinician-collected (81.8%) or home self-collected (76.1%) HPV testing was high, if recommended by a provider. Acceptability of clinician-collected HPV testing was associated with higher income ( p = 0.009) and for self-collection was associated with higher income ( p = 0.002) and higher education ( p = 0.02). Higher education was associated with reporting self-collection as easier than clinic-collection ( p = 0.02). Women expected self-collection to be more convenient (94%), less embarrassing (85%), easier (85%), and less painful (81%) than clinician-collection. Conclusions Educational interventions are needed to address low awareness about the current clinician-collected primary HPV screening option and to prepare for anticipated federal licensure of self-collection kits. Informing women about self-collection allows them to recognize benefits which could address screening barriers.

Abstract Background Error analysis plays a crucial role in clinical concept extraction, a fundamental subtask within clinical natural language processing (NLP). The process typically involves a manual review of error types, such as contextual and linguistic factors contributing to their occurrence, and the identification of underlying causes to refine the NLP model and improve its performance. Conducting error analysis can be complex, requiring a combination of NLP expertise and domain-specific knowledge. Due to the high heterogeneity of electronic health record (EHR) settings across different institutions, challenges may arise when attempting to standardize and reproduce the error analysis process. Objectives This study aims to facilitate a collaborative effort to establish common definitions and taxonomies for capturing diverse error types, fostering community consensus on error analysis for clinical concept extraction tasks. Materials and Methods We iteratively developed and evaluated an error taxonomy based on existing literature, standards, real-world data, multisite case evaluations, and community feedback. The finalized taxonomy was released in both .dtd and .owl formats at the Open Health Natural Language Processing Consortium. The taxonomy is compatible with several different open-source annotation tools, including MAE, Brat, and MedTator. Results The resulting error taxonomy comprises 43 distinct error classes, organized into 6 error dimensions and 4 properties, including model type (symbolic and statistical machine learning), evaluation subject (model and human), evaluation level (patient, document, sentence, and concept), and annotation examples. Internal and external evaluations revealed strong variations in error types across methodological approaches, tasks, and EHR settings. Key points emerged from community feedback, including the need to enhancing clarity, generalizability, and usability of the taxonomy, along with dissemination strategies. Conclusion The proposed taxonomy can facilitate the acceleration and standardization of the error analysis process in multi-site settings, thus improving the provenance, interpretability, and portability of NLP models. Future researchers could explore the potential direction of developing automated or semi-automated methods to assist in the classification and standardization of error analysis.

Abstract The Voyage Study is a large-scale, prospective cohort study of long-term health outcomes among participants with a valid multi-target stool DNA (mt-sDNA) test order. Recently, the Voyage Study achieved its enrollment target of 150,000 participants. The intent of this descriptive analysis is to provide baseline geographic and socio-demographic characteristics of Voyage Study participants to set the stage for future comparisons to the broader mt-sDNA user population. Participants were randomly selected from all persons in the U.S. or Puerto Rico who had a valid order for mt-sDNA screening during the enrollment period (10/18/2019 - 10/17/2023). Enrolled participants completed a baseline questionnaire that collected socio-demographic and health information. Participant characteristics and baseline survey answers were compared between regions using chi-square tests where appropriate. Regional geographic data was aggregated for the broader mt-sDNA user population during the enrollment period, though no formal comparisons were made between the broader population and the Voyage population at this time. Analyses were limited to 139,369 participants with available baseline questionnaire data. Voyage participants were more likely to reside in the Midwest, Northeast, or Southeast (29.9%, 22.5%, 22.6% respectively) and less likely to reside in the Southwest and West (11.6%, 13.4%). During the enrollment period, differential participation was also observed in the mt-sDNA user population, with the Midwest, Northeast, and Southeast having the greatest proportion (25.4%, 22.2%, 26.7%) and the Southwest and West having the lowest proportion of mt-sDNA users (13.8%, 11.9%). In the Voyage cohort, Midwest participants reported the lowest education attainment as characterized by years of school completed (14.1% high school or below, 59.2% college or greater), while participants in the West reported the highest education attainment (7.9% high school or below, 67.3% college or greater). The West also had the highest percentage of participants reporting an income of $100,000 or greater (51.6%), with the Southeast being the lowest (42.2%), as compared to the other regions. We report here early results of geographic and socio-demographic characteristics from a new research resource that will enable future analyses of health outcomes and behaviors in a population of mt-sDNA users. We observed that the greatest percentage of Voyage participants resided in the Midwest, Northeast, or Southeast, and that there was differential socio-demographic participant characteristics based on region. Future analyses will compare long-term healthcare outcomes and differences in colorectal cancer incidence and mortality between the broader mt-sDNA user population and the Voyage cohort. Citation Format: Allison M. Berry, Jessica A. Grimm, Christine R. Kirt, Emily J. Kirsch, Rachel E. Dixon, Winnie Z. Fan, Kandace A. Bohn, Jennifer St. Sauver, Kathleen J. Yost, Lila J. Finney Rutten, Janet E. Olson. U.S. Geographic and socio-demographic characteristics of Voyage Study participants [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 3453.

Abstract Background Body composition can be accurately quantified from abdominal computed tomography (CT) exams and is a predictor for the development of aging-related conditions and for mortality. However, reference ranges for CT-derived body composition measures of obesity, sarcopenia, and bone loss have yet to be defined in the general population. Methods We identified a population-representative sample of 4 900 persons aged 20 to 89 years who underwent an abdominal CT exam from 2010 to 2020. The sample was constructed using propensity score matching an age and sex stratified sample of persons residing in the 27-county region of Southern Minnesota and Western Wisconsin. The matching included race, ethnicity, education level, region of residence, and the presence of 20 chronic conditions. We used a validated deep learning based algorithm to calculate subcutaneous adipose tissue area, visceral adipose tissue area, skeletal muscle area, skeletal muscle density, vertebral bone area, and vertebral bone density from a CT abdominal section. Results We report CT-based body composition reference ranges on 4 649 persons representative of our geographic region. Older age was associated with a decrease in skeletal muscle area and density, and an increase in visceral adiposity. All chronic conditions were associated with a statistically significant difference in at least one body composition biomarker. The presence of a chronic condition was generally associated with greater subcutaneous and visceral adiposity, and lower muscle density and vertebrae bone density. Conclusions We report reference ranges for CT-based body composition biomarkers in a population-representative cohort of 4 649 persons by age, sex, body mass index, and chronic conditions.

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.

AbstractBackgroundPhysical activity can improve the physical health of people living with mild cognitive impairment (MCI) and dementia and may have cognitive benefits. Identifying modifiable social factors inhibiting physical activity among this group is important. We examined the relationship between reported physical activity levels and social determinants of health (SDOH) in a population of older adults living with MCI or dementia.MethodWe included people with a diagnosis of MCI or dementia, age 55 years and older, who were empaneled to Community Internal Medicine at Mayo Clinic (Rochester, Minnesota), had a clinic visit between June 1, 2019 and June 30, 2021 and had completed a SDOH questionnaire. We focused on 8 SDOH domains: education, depression, alcohol use, stress, financial resource strain, social connections, food insecurity and transportation needs. Self‐reported activity levels were categorized as inactive (0 minutes of exercise per week), insufficiently active (10‐140 minutes per week) or sufficiently active (≥150 minutes per week). SDOH domains were compared according to physical activity level using χ2 tests and multinomial logistic regression.ResultA total of 3224 persons with MCI (n = 1371) or dementia (n = 1853) who had completed questions on physical activity were included. Of these, 1936 (60%) were characterized as physically inactive and 837 (26%) insufficiently active. Characteristics associated with an increased likelihood of physical inactivity were older age, female sex, obesity, lower education, dementia diagnosis, depression risk and increased social isolation (all p<0.001).ConclusionPhysical inactivity is common among people living with MCI and dementia. Physical activity levels may be influenced by many modifiable factors, highlighting potential areas for intervention in this population.