Community engagement (CE) is critical for research on the adoption and use of assistive technology (AT) in many populations living in resource-limited environments. Few studies have described the process that was used for engaging communities in AT research, particularly within low-income communities of older Hispanic with disabilities where limited access, culture, and mistrust must be navigated. We aimed to identify effective practices to enhance CE of low-income Hispanic communities in AT research.

The community stakeholders included community-based organizations, the community healthcare clinic, the local AT project, and residents of the Caño Martín Peña Community in San Juan, Puerto Rico. The CE procedures and activities during the Planning the Study Phase comprised working group meetings with stakeholders to cocreate the funding proposal for the study and address the reviewers’ critiques. During the Conducting the Study Phase, we convened a Community Advisory Board to assist in the implementation of the study. During the Disseminating the Study Results Phase, we developed and implemented plans to disseminate the research results.

We identified seven distinct practices to enhance CE in AT research with Hispanic communities: (1) early and continuous input; (2) building trusting and warm relationships through personal connections; (3) establishing and maintaining presence in the community; (4) power sharing; (5) shared language; (6) ongoing mentorship and support to community members; and (7) adapting to the changing needs of the community.

Greater attention to CE practices may improve the effectiveness and sustainability of AT research with low-income communities.

Lack of participation in clinical trials (CTs) is a major barrier for the evaluation of new pharmaceuticals and devices. Here we report the results of the analysis of a dataset from ResearchMatch, an online clinical registry, using supervised machine learning approaches and a deep learning approach to discover characteristics of individuals more likely to show an interest in participating in CTs.

We trained six supervised machine learning classifiers (Logistic Regression (LR), Decision Tree (DT), Gaussian Naïve Bayes (GNB), K-Nearest Neighbor Classifier (KNC), Adaboost Classifier (ABC) and a Random Forest Classifier (RFC)), as well as a deep learning method, Convolutional Neural Network (CNN), using a dataset of 841,377 instances and 20 features, including demographic data, geographic constraints, medical conditions and ResearchMatch visit history. Our outcome variable consisted of responses showing specific participant interest when presented with specific clinical trial opportunity invitations (‘yes’ or ‘no’). Furthermore, we created four subsets from this dataset based on top self-reported medical conditions and gender, which were separately analysed.

The deep learning model outperformed the machine learning classifiers, achieving an area under the curve (AUC) of 0.8105.

The results show sufficient evidence that there are meaningful correlations amongst predictor variables and outcome variable in the datasets analysed using the supervised machine learning classifiers. These approaches show promise in identifying individuals who may be more likely to participate when offered an opportunity for a clinical trial.

The extent to which Clinical and Translational Science Award (CTSA) programs offer publicly accessible online resources for training in community-engaged research (CEnR) core competencies is unknown. This study cataloged publicly accessible online CEnR resources from CTSAs and mapped resources to CEnR core competency domains.

Following a search and review of the current literature regarding CEnR competencies, CEnR core competency domains were identified and defined. A systematic review of publicly accessible online CEnR resources from all 64 current CTSAs was conducted between July 2018 and May 2019. Resource content was independently reviewed by two reviewers and scored for the inclusion of each CEnR core competency domain. Domain scores across all resources were assessed using descriptive statistics.

Eight CEnR core competency domains were identified. Overall, 214 CEnR resources publicly accessible online from 35 CTSAs were eligible for review. Scoring discrepancies for at least one domain within a resource initially occurred in 51% of resources. “CEnR methods” (50.5%) and “Knowledge and relationships with communities” (40.2%) were the most frequently addressed domains, while “CEnR program evaluation” (12.1%) and “Dissemination and advocacy” (11.2%) were the least frequently addressed domains. Additionally, challenges were noted in navigating CTSA websites to access CEnR resources, and CEnR competency nomenclature was not standardized.

Our findings guide CEnR stakeholders to identify publicly accessible online resources and gaps to address in CEnR resource development. Standardized nomenclature for CEnR competency is needed for effective CEnR resource classification. Uniform organization of CTSA websites may maximize navigability.

Clinical trial participation among US Hispanics remains low, despite a significant effort by research institutions nationwide. ResearchMatch, a national online platform, has matched 113,372 individuals interested in participating in research with studies conducted by 8778 researchers. To increase accessibility to Spanish speakers, we translated the ResearchMatch platform into Spanish by implementing tenets of health literacy and respecting linguistic and cultural diversity across the US Hispanic population. We describe this multiphase process, preliminary results, and lessons learned.

Translation of the ResearchMatch site consisted of several activities including: (1) improving the English language site’s reading level, removing jargon, and using plain language; (2) obtaining a professional Spanish translation of the site and incorporating iterative revisions by a panel of bilingual community members from diverse Hispanic backgrounds; (3) technical development and launch; and (4) initial promotion.

The Spanish language version was launched in August 2018, after 11 months of development. Community input improved the initial translation, and early registration and use by researchers demonstrate the utility of Spanish ResearchMatch in engaging Hispanics. Over 12,500 volunteers in ResearchMatch self-identify as Hispanic (8.5%). From August 2018 to March 2020, 162 volunteers registered through the Spanish language version of ResearchMatch, and over 500 new and existing volunteers have registered a preference to receive messages about studies in Spanish.

By applying the principles of health literacy and cultural competence, we developed a Spanish language translation of ResearchMatch. Our multiphase approach to translation included key principles of community engagement that should prove informative to other multilingual web-based platforms.

The Clinical and Translational Science Awards (CTSA) program of the National Center for Advancing Translational Sciences (NCATS) seeks to improve population health by accelerating the translation of scientific discoveries in the laboratory and clinic into practices for the community. CTSAs achieve this goal, in part, through their pilot project programs that fund promising early career investigators and innovative early-stage research projects across the translational research spectrum. However, there have been few reports on individual pilot projects and their impacts on the investigators who receive them and no studies on the long-term impact and outcomes of pilot projects.

The Georgia CTSA funded 183 pilot projects from 2007 to 2015. We used a structured evaluation framework, the payback framework, to document the outcomes of 16 purposefully-selected pilot projects supported by the Georgia CTSA. We used a case study approach including bibliometric analyses of publications associated with the selected projects, document review, and investigator interviews.

These pilot projects had positive impact based on outcomes in five “payback categories”: (1) knowledge; (2) research targeting, capacity building, and absorption; (3) policy and product development; (4) health benefits; and (5) broader economic benefits.

Results could inform our understanding of the diversity and breadth of outcomes resulting from Georgia CTSA-supported research and provide a framework for evaluating long-term pilot project outcomes across CTSAs.

To identify phenotypes of type 1 diabetes based on glucose curves from continuous glucose-monitoring (CGM) using functional data (FD) analysis to account for longitudinal glucose patterns. We present a reliable prediction model that can accurately predict glycemic levels based on past data collected from the CGM sensor and real-time risk of hypo-/hyperglycemic for individuals with type 1 diabetes.

A longitudinal cohort study of 443 type 1 diabetes patients with CGM data from a completed trial. The FD analysis approach, sparse functional principal components (FPCs) analysis was used to identify phenotypes of type 1 diabetes glycemic variation. We employed a nonstationary stochastic linear mixed-effects model (LME) that accommodates between-patient and within-patient heterogeneity to predict glycemic levels and real-time risk of hypo-/hyperglycemic by creating specific target functions for these excursions.

The majority of the variation (73%) in glucose trajectories was explained by the first two FPCs. Higher order variation in the CGM profiles occurred during weeknights, although variation was higher on weekends. The model has low prediction errors and yields accurate predictions for both glucose levels and real-time risk of glycemic excursions.

By identifying these distinct longitudinal patterns as phenotypes, interventions can be targeted to optimize type 1 diabetes management for subgroups at the highest risk for compromised long-term outcomes such as cardiac disease or stroke. Further, the estimated change/variability in an individual’s glucose trajectory can be used to establish clinically meaningful and patient-specific thresholds that, when coupled with probabilistic predictive inference, provide a useful medical-monitoring tool.

Although organizational climate may affect faculty’s mentoring behaviors, there has not been any way to measure that climate. The purpose of this study was to test the reliability and validity of two novel scales to measure organizational mentoring climate importance and availability at two public research universities.

We developed 36 content-valid mentoring climate items in four dimensions: Structure, Programs/Activities, Policies/Guidelines, and Values. In total, 355 faculty completed an anonymous, structured, online survey asking about the importance (very important to very unimportant) and availability (no, don’t know, yes) of each of the items. We conducted reliability analyses and construct validity testing using exploratory common factor analysis, principal axis factoring, and oblique rotation.

The majority of the predominantly female, White non-Hispanic, senior, tenure-track faculty were not currently mentoring another faculty or being mentored. Analyses demonstrated a 15-item solution for both the Organizational Mentoring Climate Importance (OMCI) and the Availability (OMCA) Scales, with three factors each: Organizational Expectations, Mentor–Mentee Relationships, and Resources. Standardized Cronbach alphas ranged from 0.74 to 0.90 for the subscales, and 0.94 (OMCI) and 0.87 (OMCA) for the full scales. Faculty rated all items as somewhat to very important; however, perceived availability was very low ranging from mentor training programs (40%) to guidelines for evaluating mentoring success or managing conflict (2.5%).

The scales will allow studying of how organizational climate may affect mentoring behavior and whether climate can be changed to improve faculty mentoring outcomes. We provide recommendations for furthering the science of organizational mentoring climate and culture.

Urothelial carcinoma (UCC) develops in both humans and dogs and tracks to regions of high industrial activity. We hypothesize that dogs with UCC may act as sentinels for human urothelial carcinogen exposures. The aim of this pilot study was to determine whether healthy people and dogs in the same households share urinary exposures to potentially mutagenic chemical carcinogens.

We measured urinary concentrations of acrolein (as its metabolite 3-HPMA), arsenic species, 4-aminobiphenyl, and 4-chlorophenol (a metabolite of the phenoxyherbicide 2,4-D) in healthy dogs and their owners. We assessed possible chemical sources through questionnaires and screened for urothelial DNA damage using the micronucleus assay.

Biomarkers of urinary exposure to acrolein, arsenic, and 4-chlorophenol were found in the urine of 42 pet dogs and 42 owners, with 4-aminobiphenyl detected sporadically. Creatinine-adjusted urinary chemical concentrations were significantly higher, by 2.8- to 6.2-fold, in dogs compared to humans. Correlations were found for 3-HPMA (r = 0.32, P = 0.04) and monomethylarsonic acid (r = 0.37, P = 0.02) between people and their dogs. Voided urothelial cell yields were inadequate to quantify DNA damage, and questionnaires did not reveal significant associations with urinary chemical concentrations.

Healthy humans and pet dogs have shared urinary exposures to known mutagenic chemicals, with significantly higher levels in dogs. Higher urinary exposures to acrolein and arsenic in dogs correlate to higher exposures in their owners. Follow-up studies will assess the mutagenic potential of these levels in vitro and measure these biomarkers in owners of dogs with UCC.

Food insecurity (FI) is the lack of consistent access to enough food for an active and healthy life. Community-based hunger relief programs often serve as emergency food sources for families with FI. However, these programs may not provide foods that diverse populations of people prefer. We sought to evaluate the dietary patterns and preferences of families living in food-insecure neighborhoods and utilizing a community-based hunger relief program, in order to improve the utilization of local nutritional programs.

We examined the Help Our People Eat (HOPE) community-based mobile meal program. Free-listing interviews (n = 63) were conducted with English-(66%) and Spanish-speaking (34%) participants of the program. Participants were asked about FI risk, food preferences, and dietary behaviors at home.

The majority of participants (90%) had children in the household. About 60% reported not being able to afford the type of food they enjoyed. Most participants reported using stoves for cooking (80%). Participants overwhelmingly cooked with chicken, beef, and pork. The most common side dishes included potatoes, rice, and salad. Most participants reported no interest in cooking differently or learning new recipes.

A common theme throughout interviews was that families prefer similar meals, but may prepare them differently based on the language spoken. Food preferences consisted of a high intake of carbohydrate-rich meals, perhaps because these foods may be cheaper and easier to access. Notably, new recipes and cooking methods were not a priority for these families, possibly due to the time and effort needed to learn them.

Access to patient medical data is critical to building a real-time data analytic pipeline for improving care providers’ ability to detect, diagnose, and prognosticate diseases. Critical congenital heart disease (CCHD) is a common group of neonatal life-threatening defects that must be promptly diagnosed to minimize morbidity and mortality. CCHD can be diagnosed both prenatally and postnatally. However, despite current screening practices involving oxygen saturation analysis, timely diagnosis is missed in approximately 900 infants with CCHD annually in the USA and can benefit from increased data processing capabilities. Adding non-invasive perfusion measurements to oxygen saturation data can improve the timeliness and fidelity of CCHD diagnostics. However, real-time monitoring and interpretation of non-invasive perfusion data are currently limited.

To address this challenge, we created a hardware and software architecture utilizing a Pi-top™ for collecting, visualizing, and storing dual oxygen saturation, perfusion indices, and photoplethysmography data. Data aggregation in our system is automated and all data files are coded with unique study identifiers to facilitate research purposes.

Using this system, we have collected data from 190 neonates, 130 presumably without and 60 with congenital heart disease, in total comprising 1665 min of information. From these data, we are able to extract non-invasive perfusion features such as perfusion index, radiofemoral delay, and slope of systolic rise or diastolic fall.

This data collection and waveform analysis is relatively inexpensive and can be used to enhance future CCHD screening algorithms.

Identifying predictors of patient outcomes evaluated over time may require modeling interactions among variables while addressing within-subject correlation. Generalized linear mixed models (GLMMs) and generalized estimating equations (GEEs) address within-subject correlation, but identifying interactions can be difficult if not hypothesized a priori. We evaluate the performance of several variable selection approaches for clustered binary outcomes to provide guidance for choosing between the methods.

We conducted simulations comparing stepwise selection, penalized GLMM, boosted GLMM, and boosted GEE for variable selection considering main effects and two-way interactions in data with repeatedly measured binary outcomes and evaluate a two-stage approach to reduce bias and error in parameter estimates. We compared these approaches in real data applications: hypothermia during surgery and treatment response in lupus nephritis.

Penalized and boosted approaches recovered correct predictors and interactions more frequently than stepwise selection. Penalized GLMM recovered correct predictors more often than boosting, but included many spurious predictors. Boosted GLMM yielded parsimonious models and identified correct predictors well at large sample and effect sizes, but required excessive computation time. Boosted GEE was computationally efficient and selected relatively parsimonious models, offering a compromise between computation and parsimony. The two-stage approach reduced the bias and error in regression parameters in all approaches.

Penalized and boosted approaches are effective for variable selection in data with clustered binary outcomes. The two-stage approach reduces bias and error and should be applied regardless of method. We provide guidance for choosing the most appropriate method in real applications.