Kate Winskell, PhD
Associate Professor, Hubert Department of Global Health
Rollins School of Public Health, Emory University

Leveraging technology to address diabetes disparities: User-centered design of a smartphone app to improve glycemic control through behavioral modification among CGM users.

Project Summary:

This proposal seeks to contribute to reductions in diabetes disparities by optimizing the impact, reach, and sustainability of (a) an efficacious technology, continuous glucose monitoring (CGM), and (b) an efficacious behavioral strategy, diabetes self-management education and support (DSMES), among populations vulnerable to some of the worst diabetes outcomes. Our objective is to employ iterative user-centered design approaches with patients, providers, and other stakeholders at Grady Health System to co-create a theory-based smartphone application to improve glycemic control through behavioral modification among underserved patients using CGM.

Our proposal bridges the three GCDTR cores and builds on preliminary research with Grady patients and providers to establish project feasibility and identify content and design needs. EGDRC has provided $20k in seed funds to support the development of the app by Gari Clifford’s lab (Emory School of Medicine). In order to ensure that these funds are optimally expended, it is essential that we engage stakeholders in the proposed iterative, participatory design process to inform and test evolving content, wireframes, and prototype components. This proposal, if funded, will therefore allow us to build the app and thereby support an application to NIDDK’s small R01 program to feasibility test it.

Project Goals:

To engage patients (n=16), providers (n=12), and other stakeholders (n=8) in an iterative user-centered design process to inform and test evolving content, wireframes, and prototype components.

To conduct preliminary usability testing of the app with Grady patients (n=6).

Leslie Johnson, MPH, PhD
Assistant Professor, Department of Family and Preventive Medicine
Emory University School of Medicine

Assessing and Identifying Strategies to Reduce the Postpartum Diabetes Screening Gap among Women with a History of Gestational Diabetes Mellitus in Georgia

Project Summary:

Gestational diabetes mellitus (GDM) is a common complication of pregnancy associated with a seven-fold increase in lifetime risk for diabetes. Poor adherence to postpartum diabetes screening recommendations mitigates the ability to effectively identify at-risk individuals and intervene to reduce the risk of developing T2DM. In the U.S., diabetes screenings within the recommended twelve weeks postpartum have been shown to occur anywhere between 3.4% to 57.9% of the time. Georgia has the highest rate of gestational diabetes in the southeastern United States and offers an opportune setting to study this notable gap in implementation of postpartum diabetes screening.

The goal of this study is to measure the postpartum diabetes screening gap in Georgia and explore barriers to and facilitators of postpartum diabetes screening to identify potential strategies to improve postpartum screening rates. We will interview racially/ethnically diverse women with a history of GDM to understand their healthcare experiences and preferences for diabetes screening in the postpartum period. We will also interview health care providers to understand clinician-barriers to postpartum diabetes screening and perceived needs to increase screening practices and linkage to other care providers. The investigators will use data collected from this GCDTR supported project to guide future implementation studies.

Project Goals:

1. Measure the postpartum diabetes screening gap among women with a history of gestational diabetes mellitus (GDM).
2. Explore barriers to and facilitators of postpartum diabetes screening to identify potential strategies to improve postpartum screening rates in Georgia.

Shakira Sugila, ScD
Professor, Department of Epidemiology
Rollins School of Public Health, Emory University

The Diabetes Prevention Maternal Study: a pilot home environment intervention for new moms

Project Summary:

Behavioral interventions delivered during sensitive life periods, such as the maternal postpartum phase, may be the most effective in reducing risk of Type 2 Diabetes (T2DM). Few studies have examined how families and home environments impact T2DM and cardiometabolic health. Focusing on the home environment may be a novel and potentially fruitful approach. We propose to adapt the established Health Homes/Healthy Families (HH/HF) intervention to target, diet and physical activity through the home environment, starting early in the postpartum period. In this pilot study, we aim to identify psychosocial and environmental determinants of nutrition and physical activity behaviors for T2DM prevention among new mothers from interactions with providers and postpartum women; adapt HH/HF nutrition, physical activity and weight management content for delivery to new mothers; and obtain feedback on intervention strategies and delivery from the target population. We will partner with the Georgia Department of Public Health, Home Visiting Program to facilitate interviews with providers and postpartum women. The next phase funding will test how this adapted intervention, delivered in conjunction with evidence-based home visiting programs, impacts maternal cardiometabolic health in a randomized control trial. This study has the potential to inform T2DM prevention among a medically underserved population.

Project Goals:

1. Identify psychosocial and environmental determinants of nutrition and physical activity behaviors for T2DM prevention among new mothers, from interactions with providers and existing literature.
2. Identify psychosocial and environmental determinants of nutrition and physical activity behaviors for T2DM prevention among new mothers, from interactions with postpartum women.
3. Adapt content from research-tested home environment healthy lifestyle interventions targeting healthy eating, increased physical activity and weight management, for delivery to new mothers.

Shihao Yang, PhD
Assistant Professor, H. Milton Stewart School of Industrial & Systems Engineering
Georgia Institute of Technology

Longitudinal treatment outcomes for diabetes-diagnosed youth and young adults using Medicaid claims

Project Summary:

The increasing prevalence of both type 1 diabetes (T1DM) and type 2 diabetes (T2DM) in youth and young adults is a growing concern. We propose a retrospective longitudinal study using Medicaid claims data to analyze healthcare utilization in youth and young adults with diabetes, with focus on southeast states. We will first identify the prevalence of T1DM and T2DM diagnosis and demographic characteristics of Medicaid-insured youth and young adults receiving diabetes care in the US southeast, particularly how it has changed over a period of 14 years (2005-2018). We will then use statistical models to assess patterns in healthcare services. In particular, we will examine the healthcare utilization rate of diabetes disease management vs severe outcomes. To better understand the disparities across demographic factors, we plan to stratify the analysis by year, state/county, rurality/urbanicity, age groups, and race/ethnicity. Finally, a novel causal inference framework will be applied to study the causal effect of diabetes disease management on severe outcomes, and identify the population groups where such effect is particularly strong or weak. The results of this study will better inform health agencies in addressing the gaps in care for the Medicaid population, especially on the potential gaps in healthcare delivery by various population strata.

Project Goals:

• Identify the prevalence and demographic characteristics of T1DM and T2DM among Medicaid users in the US Southeast and how it has changed over a period of 14 years (2005-2018).
• Assess patterns in health services use among youth and young adults with T1DM and T2DM.
  • To determine the type of health service utilization.
  • To assess longitudinal healthcare utilization by geographic region of health services utilization among youth adult Medicaid users with diabetes.
  • To draw inferences in systematic disparities in disease management severe outcomes with respect to geographic and individual factors.

Jessica Harding, PhD
Assistant Professor, Department of Surgery
Emory University School of Medicine  

Understanding the Impact of COVID-19 on Diabetes Care and Management Risk: An Interrupted Time-Series Analysis

 Project Summary: 

Accumulating evidence suggests that people with diabetes are more likely to have severe or fatal COVID-19 compared with people without diabetes. However, conclusions drawn from such studies are hampered by inherent selection biases of hospitalizations populations, small sample sizes, limited covariate data and no longitudinal follow-up. The extent to which diabetes confers a real and clinically meaningful increased risk forCOVID-19 and COVID-19-related outcomes (both in the short and long-term) is unknown due to alack of robust, diverse and multi-regional population-based data with longitudinal follow-up necessary to understand COVID-19 in the general diabetes population.   

Furthermore,  a  population-based  COVID-19  risk  score  specific  to  the  diabetes  population  is  yet  to  be developed. The use of novel machine learning approaches in large existing registry-based electronic health records (EHR) offers a unique and timely opportunity to develop a COVID-19 risk score to identify individuals with diabetes who will most likely benefit from tailored interventions. In this proposal we will leverage the extensive Veterans Affairs (VA) and Kaiser Permanente Georgia (KPGA) virtual  data warehouses (VDW) with  data  on  more than  4  million and  330,00  patients,  respectively, to:  1) provide robust population-based estimates of COVID-19 and COVID-19-related outcomes in people with vs. without type 2 diabetes(T2D); and 2) develop and validate a COVID-19 risk score in people with T2D, using machine  learning  techniques  that incorporated demographic,  clinical,  geographic,  hospitalization,  and socioeconomic information. Results from this study will provide much needed population-based estimates on risk of COVID-19 people with T2D needed to inform current clinical and public health decision-making and resource allocation. Further, the risk score generated from this pilot project will inform subsequent research funding applications that focus on a) expanding  data  collection;  b)  generating  a  diabetes  and  COVID-19  cohort;  and  c)  designing  and implementing novel interventions that mitigate the impact of COVID-19 in people with T2D. Collectively, this proposal has the potential to make meaningful and translatable impacts for people with T2D suffering from COVID-19 and its sequelae.   

Project Goals:  

1. Determine the incidence of COVID-19 and the short and long-term COVID-19-related outcomes, in people with vs. without type 2 diabetes. 
2. Develop and validate a risk score, using novel machine learning techniques, to predict risk of COVID-19 and COVID-19-related outcomes in adults with type 2 diabetes. 

Andrea G. Parker, PhD
Associate Professor, School of Interacting Computing
Georgia Institute of Technology

Establishing a technology ecosystem to optimize COVID-19 testing

Project Summary:

This pilot project aims to jumpstart a technology ecosystem to iteratively optimize the design, deployment, and evaluation of COVID-19 testing for GA residents who are at risk for, or are diagnosed with diabetes. This approach will be responsive to the dynamics of observed and predicted COVID-19 disease spread, the vulnerability of Georgia populations with or at risk for diabetes, and behavioral, educational, and attitudinal COVID-19 testing barriers.

Project Goals:

Optimize the targeted deployment of testing resources, dynamically adjusted geographically and over time, based on agent-based predictive models for COVID-19 spread and simulation models of testing capacity.

Optimize the testing experience for staff and recipients, including messaging for testing recruitment based on media and social media monitoring, and providing usable testing and educational materials.

Francisco Pasquel, MD, MPH
Associate Professor, Department of Medicine
Emory University School of Medicine

The Preparedness Cycle in the Care of People with Diabetes During the COVID-19 Pandemic and Future Outbreaks: Mitigation and Surveillance Strategies to evaluate changes in diabetes care

Project Summary:

The COVID-19 pandemic is being considered as a mass casualty incident (MCI) of the highest nature.4 As of April 17, 2020 more than two million people have tested positive for COVID-19 globally. The US has the highest number of cases in the world and thus far more than 30,000 people have died. It is expected that many more people will get the infection during the peak of the pandemic and potential new outbreaks are probable after the lockdown alleviates.5 A large proportion of patients with COVID-19 have diabetes, which has been recognized as a common comorbidity among those with severe disease.6  In this proposal we aim to evaluate components of the Preparedness Cycle framework (utilized in emergency/disasters) to immediately respond to this public health crisis and mitigate potential harms related to practice change or clinical inertia associated with fear and a lack of PPE. We will also assess failures and challenges in the care of patients with diabetes in the time surrounding this pandemic and design a plan to follow patients after the crisis in preparation for similar situations in the future, (Figure). We have identified four areas that require attention during this pandemic with the goal of informing the community about gaps in care at this time and potential guidance to prepare for future outbreaks:

• Mitigation (act now): we will evaluate ongoing Quality Improvement (QI) efforts and implement diabetes technology for remote continuous glucose monitoring (CGM) and individualized inpatient diabetes management to reduce PPE waste and healthcare workers (HCW) exposures.
• Surveillance (monitor and predict achievement of goals of care and outcomes): continuous, systematic collection, analysis and interpretation of the diabetes continuum of care in patients with diabetes with and without COVID-19 during the current pandemic and future outbreaks
• Preparedness (how prepared were we?, prepare for the next outbreak): we will conduct a qualitative assessment of urgent adaptations of outpatient and inpatient care delivered during the COVID-19 pandemic to care for people with diabetes. We will explore gaps and successes from provider and patient perspectives.
• Recovery phase (after the pandemic): Assessment of achievement of diabetes care benchmarks according to diabetes and cardiovascular risk reduction guidelines before and after the lockdown to evaluate the impact of the pandemic on diabetes care metrics.

Project Goals:

1. Hospital integration of technology to care for patients with diabetes during the COVID -19 pandemic.
2. To implement two databases to evaluate inpatient and outpatient outcomes in patients with diabetes before, during and after the COVID-19 pandemic.

Jean Welsh, PhD, MPH, RN
Associate Professor, Department of Pediatrics
Emory University School of Medicine

Strengthening and Supporting the Role of Primary Healthcare Providers in Diabetes Early Prevention Efforts

Project Summary:

The prevalence of diabetes has risen markedly among U.S. children, greatly increasing their lifetime risk of a decreased quality of life and premature mortality. Numerous well-designed studies have highlighted early eating patterns associated with elevated risk, including the consumption of sugar-containing beverages (SCBs). SCB consumption among U.S. children begins early, for many before the age of 1 year. While consumption in infancy tends to be in the form of naturally sweet fruit juice rather than the sugar-sweetened beverages, the similarity in the sugar and calorie content of these beverages suggests that their impact when consumed would be similar. Though limited, studies examining the impact of SCBs in infancy suggest that those who consume sugar-sweetened beverage and fruit juices, alone or in combination, before their first birthday are more likely to be overweight or obese and at increased risk for developing diabetes as they age, than infants who do not. Pediatric primary healthcare providers are well-positioned to provide SCB reduction related education and counseling for parents of infants and young children as part of routine well-child visits. Unfortunately, though, limited time, training in nutrition, and the need address many competing demands, all combine to limit the frequency and the effectiveness of such efforts. Videos, while shown to be an effective tool for helping to ensure that key education messages are easily, accurately, and consistently conveyed, they appear to be rarely used in the pediatric primary healthcare setting. In follow-up to a recent study in which we demonstrated that incorporating educational videos into routine well-child care was well received by both parents and clinicians, we propose the following study to determine if the use of age-specific videos designed specifically to promote parent compliance with SCB are effective in: 1) Increasing parents’ SCB-related knowledge and their compliance with related early feeding recommendations and, 2) Increasing the frequency and quality of the SCB-related reduction education and counseling provided to parents during well-child visits. The results of this pilot will be used to guide the development of a larger RCT to determine if a multicomponent intervention in pediatric primary healthcare can effectively reduce children’s SCB consumption and lower their risk of diabetes.

Project Goals:

1. The goal of the proposed pilot project is to assess the impact of using targeted, age-specific videos during well-child visits as a tool for increasing parent compliance with feeding practices that promote a reduction in children’s SCB consumption. Specifically, we aim to determine if incorporating these videos into routine healthcare practice:
2. Increases parents’ SCB-related knowledge and their compliance with related early feeding recommendations. The primary outcome will be the percentage of parents, post- intervention compared to pre-intervention, who report that they do not give their infants SCBs.
3. Increases the frequency and quality of the SCB-related reduction education and counseling provided to parents during well-child visits. A secondary outcome measure will be the proportion of parents who recall receiving information during the previous well-child visit about the need to avoid giving their infant SCBs.

Karla Galaviz, PhD
Assistant Professor, Hubert Department of Global Health
Rollins School of Public Health, Emory University

Ines Gonzales-Casanova, PhD
Assistant Professor, Hubert Department of Global Health
Rollins School of Public Health, Emory University

Exploring health disparities at the intersection of HIV, diabetes and depression among Mexican immigrants

Project Summary:

HIV, diabetes and depression disproportionally affect Mexican immigrants in the US but the factors contributing to these disparities are not fully understood. This knowledge gap limits our ability to effectively detect and manage these ailments in the largest foreign-born group in the US. The unique factors driving health disparities in Mexican immigrants should be identified to better tailor preventive interventions and direct healthcare resources more cost-effectively. With the goal of identifying opportunities to address health disparities, the objective of this study is to elucidate factors associated with HIV, diabetes and depression risk, awareness and linkage to preventive care among Mexican immigrants in the Southeast. This study will be informed by a health disparities trajectory model that uses a syndemic approach specifically designed to address the various socioecological and behavioral factors that contribute to health disparities in immigrant populations. For this, we will leverage our long-term collaboration with the General Consulate of Mexico in Atlanta through the Ventanilla de Salud (Health Window) program, which is implemented across the 50 Mexican Consulates in the US and offers health education, screening, and referral services while people wait for consular services. The Health Window in Atlanta serves Mexicans from Georgia, Tennessee and Alabama and reaches ~3000 people on a monthly basis. Using a sequential mixed-methods approach, we will explore associations between socio-biological factors, risk awareness for HIV, diabetes and depression, and care seeking behaviors among adult Health Window users (Aim1),and identify socioecological factors that influence these behaviors via semi-structured interviews(Aim 2).

Project Goals:

1. Assess the relationships between socio-biological factors (measured using validated questionnaires), risk awareness for HIV, diabetes and depression (measured using validates risk-scores), and care seeking behaviors (measured by follow-up calls after receiving a referral) among adult Health Window users.
2. Identify sociocultural factors that influence capability, opportunity and motivation for HIV/diabetes/depression care seeking behaviors via semi-structured interviews.

Amy Webb Girard, PhD
Associate Professor, Hubert Department of Global Health
Rollins School of Public Health, Emory University

Stacie Schmidt, MD
Associate Professor, Department of Medicine
Emory University School of Medicine

Improving diabetes outcomes for food insecure patients: Effectiveness of the Grady Fruit and Vegetable Prescription Program

Project Summary:

In 2017, nearly one in six Georgians and one in 5 Atlantans were food insecure meaning they did not have consistent access to enough food for an active, healthy life. A 2015, study assessing the association between food insecurity and diabetes in the Grady primary care center (PCC) noted that among 338 survey respondents, more than half were at risk of food insecurity while more than one third (34%) were both food insecure and had diabetes or pre-diabetes. Research increasingly confirms a relationship between food insecurity, the development of Type II diabetes, poorer disease management and diabetes-related morbidities. Hypothesized mechanisms include diminished diet quality, disordered eating patterns and metabolic perturbations due to associated emotional and mental stress. Traditional lifestyle change interventions do not account for the challenges that food insecure patients have accessing healthier food options. The Grady Fruit and Vegetable Prescription Program (FVRX)is an innovative lifestyle change intervention implemented within the Grady Health Care Primary care network. The FVRX program aims to improve diets and health outcomes among food insecure patients at risk for diabetes and other diet-related chronic disease by reducing food access barriers and enhancing skills and self-efficacy for self-management of diet related disease. Preliminary analysis of data from 2017 and 2018 cohorts suggested improvements in BMI, blood pressure and hemoglobin A1C. However, these findings are of limited value given small sample sizes and lack of appropriate control groups. Similar limitations exist for produce prescription program evaluations in other states. To overcome these limitations, we will identify a contemporaneous matched control group using de-identified electronic medical records. Controls will be matched on timing of care, clinic site, age, gender, health status and health care utilization. We will use generalized estimating equations to compare changes in BMI, blood pressure and hemoglobin A1C over time between intervention and controls. Findings will be used to develop an application for diabetes translation research that will utilize a multi-phase optimization strategy trial (MOST) approach to refine the FVRX intervention package and test the effectiveness and cost-effectiveness of the refined package on diabetes related health outcomes using a randomized controlled design. Evidence of effectiveness and cost-effectiveness would support policy efforts to allow reimbursement for participation in FVRX programs by Medicaid, Medicare and private insurance.

While these preliminary findings are suggestive of success, sample sizes are small and the evaluation lacked a control group, limiting conclusions regarding effectiveness. Produce prescription programs in other states have similar evaluation limitations. Further, the multi-component nature of the FVRX program precludes knowing which aspects of the program are contributing to change and the benefits of their inclusion relative to their costs. These limitations in the evidence base hinder uptake and sustainability within the Grady Health System and scale up elsewhere. Our goal is to evaluate and refine the FVRX intervention to optimize health outcomes for low-income, food insecure patients at risk for diabetes while minimizing costs to implementing partners thus providing a robust evidence base for policy change.

Project Goals:

1. Provide more robust evidence on the effectiveness of the FVRX program by estimating the impact of the FVRX intervention on selected health measures compared to contemporaneous, matched controls that were eligible for but did not receive the intervention. Data from these matched controls will be extracted from Grady electronic medical records.
2. Findings from Aim 1 will help develop an application for diabetes translation research that will apply a multi-optimization strategy trial to develop an intervention package that optimizes outcomes while minimizing costs. This optimized intervention will be tested using an RCT design against an evidence-based lifestyle change intervention.

Rodolfo Galindo, MD
Associate Professor, Department of Medicine
Emory University School of Medicine

Exploring health disparities at the intersection of HIV, diabetes and depression among Mexican immigrants

Project Summary:

Diabetes mellitus (DM) is a leading cause of end-stage renal disease (ESRD), with 40% of patients on dialysis having DM. Management of DM among dialysis patients is challenging due to alterations on glucose metabolism, insulin sensitivity, and also because most glucose-lowering medications are contraindicated or require dose adjustment5.Most patients with type 2 DM (DM2) and ESRD are treated with insulin (60-70%) or with sulfonylureas (40%)or both. Notably, treatment with insulin and/or sulfonylureas (SU) is associated with a higher risk for emergency department (ED) visits and hospitalizations for hypoglycemia7,8. With attentive 21stcentury diabetes care, acute hypo-and hyperglycemic crises should be almost completely avoidable. However, in this high risk population, the prevalence and impact of different insulin formulations and non-insulin medications among the ESRD/dialysis population on the frequency of hypoglycemia and hospital admissions are poorly described at a national and regional level in the United States. Similarly, even though it is well known that African Americans have a higher burden of DM and ESRD, itis unknown if minority groups in the Southeastern have high rates of ED visits and hospitalizations for severe hypoglycemia and hyperglycemia. Accordingly, the proposed study addresses a number of key research gaps with downstream translation implications. This proposal leverages access to the most comprehensive database of patients with chronic kidney disease, the United States Renal Data System (USRDS),to characterize diabetes crises, health outcomes, and pharmacoepidemiology in a nationwide sample.

Project Goals:

1. To describe annual rates of severe hypoglycemia and hyperglycemia requiring ED visits and hospitalization in a nationwide cohort of patients with DM and ESRD from 2008-2017, presented per 1000 person-years, overall and by key patient (diabetes type, sex, race/ethnicity, and age group) and geographical covariates (South, Northeast, Midwest, West, and specific states).
2. To describe the health impact of severe hypoglycemia and hyperglycemia, we will estimate 30-day all-cause readmissions, 30-day mortality, 1-year mortality, hospital costs, and hospital length of stay following ED visits or hospitalizations for severe hypoglycemia and hyperglycemia.
3. To characterize contemporary trends in utilization of anti-diabetic medications and its relationship to each diabetic crises, presented as age and sex-standardized rates for each year, and by key drug categories: a) oral hypoglycemic agent (OHA),b) combination therapy (prescriptions filled on the same day), c) basal insulin alone or in combination with OHA, except SU, d) basal insulin in combination with SU, e)multiple daily injections (MDI) insulin regimen, f) sulfonylureas alone or in combination with other OHAs, g) basal insulin analogs, h) basal human insulin, i) continuous subcutaneous insulin infusion (CSII).

Salimah LaForce, MS
Senior Policy Analyst, Ivan Allen College of Liberal Arts
Georgia Institute of Technology

American Sign Language-Accessible Diabetes Education (ASL-ADE)

Project Summary:

According to recently published data, 25% of the U.S. population is comprised of adults with disabilities, with approximately 6% of the sample having a hearing disability. Prevalence of hearing disabilities increases with age and poverty level. These estimates are likely conservative due to (a) their limitation to non-institutionalized adults, (b) sampling method, and (c) their reliance on self-identification. People who are culturally Deaf view their deafness as a difference, and thus do not self-identify as having a disability. People who are culturally Deaf view themselves as a linguistic minority, bonded by their use of a common signed language, shared experiences, and shared values and norms (e.g., collectivism, rejection of the disability label, Deaf pride, solidarity, directness, expressiveness). Like other minority populations in the United States, people who are culturally Deaf are at risk for developing diabetes and acute complications related to unmanaged diabetes. Also, there is a diabetes subtype, maternally inherited diabetes and deafness (MIDD) that, although rare, accounts for a portion of diabetes cases in the deaf population. Increased risk is associated with low rates of health literacy due to lack of accessible health information and isolation from the hearing world. Further, because the language and communication barrier for the population is so pervasive, it is anticipated that standard transition of care materials may overestimate the diabetes knowledge of people who are Deaf and transitioning into independent care routines. As such, materials may need to be restructured to account for any knowledge deficits. This project will not only test the efficacy of an educational intervention, but the research is designed in such a way as to evaluate if there are any between-group differences in knowledge about diabetes based on whether the subject has a diabetes diagnose is or not.

Project Goals:

1. Quantitatively demonstrate a need for diabetes educational outreach materials to be accessible to people who are Deaf and rely on ASL for clear and effective communications.
2. Test an a priori hypothesis that people who are Deaf without a diabetes diagnosis and those that are Deaf with a diabetes diagnosis will have similar levels of diabetes knowledge.
3. Evaluate the effect of the educational intervention on knowledge and health behaviors.

Dayna Johnson, PhD, MPH, MSW, MS
Assistant Professor, Department of Epidemiology
Rollins School of Public Health, Emory University

Unjali Gujral, PhD, MPH
Assistant Professor, Hubert Department of Global Health
Rollins School of Public Health, Emory University

Adapting a Mindfulness Intervention to Improve Sleep and Reduce Diabetes Risk Among a Diverse Sample in Atlanta

Project Summary:

Type 2 diabetes mellitus (T2DM) is a significant public health burden that requires immediate research attention to reduce morbidity and mortality, particularly in Georgia where prevalence is high. Traditional approaches to prevent T2DM have mainly consisted of lifestyle modification (e.g. diet and physical activity) and pharmacological strategies. These approaches have largely ignored, sleep health, a novel metabolic risk factor. A growing body of literature suggests a strong link between sleep and diabetes, therefore intervening on sleep may reduce the burden of T2DM, particularly in high-risk populations. Psychosocial stress is a major contributor to poor sleep, and may be a novel intervention target. The goal of this project is to conduct a pilot study to determine the feasibility of delivering a mindfulness-based stress reduction (MBSR)plus sleep education intervention to improve sleep and subsequently improve subclinical biomarkers ofT2DM risk among a community sample of racially diverse adults (N=20) with poor sleep quality and pre-diabetes. The results of this study will directly inform a larger study to be pursued in a NIH R01 application targeted at conducting a randomized-controlled trial to test the efficacy of improving sleep and subclinical diabetes through sleep education only vs. MBSR plus sleep education intervention.

Project Goals:

1. Determine the feasibility of delivering via phone-based application, a mindfulness-based stress reduction plus sleep education intervention to improve sleep health. We hypothesize that the MBSR intervention will improve sleep quality, extend sleep duration and improve general sleep complaints.
2. Conduct qualitative research to understand the uptake and usability of the mindfulness-based stress reduction intervention for sleep improvement.
3. Assess whether improving sleep through mindfulness-based stress reduction improves subclinical biomarkers associated with T2DM risk. We hypothesize that participants who complete the MBSR plus sleep education intervention will have improvements in glucose levels, insulin sensitivity and secretion.

Douglas A. Bodner, PhD, PE
Principal Research Engineer, School of Industrial & Systems Engineering
Georgia Institute of Technology

Characterizing Diabetes Screening Systems for Design, Evaluation and Disparity Identification

Project Summary:

The U.S healthcare system is characterized by unequal access to prediabetes and diabetes screening, diagnosis, and specialized high-quality care linkages. Consequently, a large number of people are excluded from the benefits of proven interventions to delay and prevent diabetes and its serious complications. This contributes to major inequality in diabetes outcomes such that low-income and race-ethnic minority groups lag behind their socioeconomically advantaged peers. Gaps in screening and in subsequent care are driven by complex factors, such as health work force availability. These factors need to be carefully modeled and understood. Furthermore, it is unclear what constraints exist in the entire health delivery system – spanning from patients, their community resources, clinics, and higher levels of care—and what the capacity requirements are. Addressing these questions becomes critical in light of current recommendations by expert bodies to detect undiagnosed diabetes and prediabetes and link those individuals to high-quality interventions. After all, it is estimated that approximately 90% of 84 million adults have undiagnosed prediabetes, and that 15-25% of the adult population with diabetes go undiagnosed (2017 CDC National Diabetes Statistics Report). The ultimate goal of this research is to comprehensively investigate capacity for a prediabetes/diabetes screening and linked care program, and model how the different agents, and resulting capacity, may behave and impact outcomes when the interactions are varied. We propose to create and use a computational modeling/simulation approach that embraces real-world complexity to achieve this goal. Traditional system modeling approaches seek to decompose the overall problem of system performance and management into isolated component elements and subsequently recompose these by integrating the designed solution for each element into an overall system design. By modeling phenomena in isolation, analysis can result in the loss of important information about interactions among the overall phenomena of interest, and thus result in ineffective translation of evidence into implementation. Interactive complex systems modeling and visualization methods on the other hand generate features of observed networks from the bottom up, and allow iterative exploration of system and capacity constraints in a dynamic manner. Furthermore, these traditional system modeling approaches have not accounted for such phenomena as screening and care capacity and patient behavior when applied to diabetes progression and treatment interventions.

In this pilot grant proposal, we propose to develop the foundation to describe system capacity for diabetes detection and linkage to appropriate care. Specifically, we will develop a conceptual model and mapping of the diabetes screening and linked care system. The pilot work will be grounded in, and validated by, the use of best available data enhanced with specific new data collection and expert opinions from stakeholder groups. The model will use a visual analytic platform to inform and engage stakeholders in detailed questions about designing and evaluating diabetes screening and linked care, as well as identification/reduction of disparities.

Project Goals:

1. Conceptually develop a model that maps in an iterative and visual manner the influential agents involved in a diabetes screening and linkage system, identify data requirements to build a complex model, and identify best available data and identify areas for novel data collection where required. The initial model will be focused on (i) a narrow geographic area (the city of Atlanta), with the intent to develop a generalizable, research-focused model beyond after demonstrating success; (ii) research questions pertinent to understanding system capacity in relation to reducing disparities in screening (e.g., what are the major constraints to screening in predominantly African American communities, which face a higher burden of diabetes?; how would screening behaviors change if Medicaid were expanded?).
   a. Rigorously characterize data inputs to implement the model.
2. Engage a stakeholder panel of experts for their reaction and feedback to this conceptual model.
3. Implement and validate the initial model developed in Aims 1 and 2 to illustrate how the model can be used to design screening programs and address disparities.

Jennifer Frediani, PhD, RD, ACSM-CES
Assistant Professor, Nell Hodgson Woodruff School of Nursing
Emory University

Feasibility of a Soccer-based Adaptation of the Diabetes Prevention Program

Project Summary:

The Diabetes Prevention Program (DPP) has been successful in reducing the incidence of T2DM through achievement of moderate weight loss and physical activity (PA) goals.(6) Evidence from the original DPP cohort (DPP Outcome Study) suggests that adoption of increased PA habits are maintained up to 10 years among intervention participants when compared to similar adults in the general population.(7) Further, cultural adaptation and implementation studies show positive results in translating the DPP to better serve specific populations using community-based resources.(8, 9)

Currently 48% of U.S. adults and 24% of Hispanics in Georgia do not meet the 2008 Federal PA guideline of 150 minutes/week of moderate-to-vigorous PA.(10) As an alternative to traditional exercise programs, recreational sport provides a unique environment that may lead to increased PA participation and motivation.(11) Results from a recent meta-analysis by our group suggest that sport-based interventions, particularly soccer, lead to significant improvements in weight, waist circumference, percentage body fat, lipid profiles, blood pressure and measures of fitness, with a greater change in males versus females. (12)

Our goal for this proposal is to leverage the local Hispanic community’s interest in soccer to deliver an evidence-based, T2DM prevention program tailored to Latino males (35-55 years) at high risk for T2DM.

Project Goals:

1. To adapt the current National Diabetes Prevention Program (NDPP) curriculum to a novel, translational soccer-based DPP programming and delivered by trained coaches as facilitators.
2. To conduct formative research addressing barriers and acceptability for implementation of this soccer-based DPP adaptation among Latino adults via questionnaires and key informant interviews.
3. To assess implementation feasibility of this novel soccer-based DPP adaptation by conducting a single-arm, pre-post pilot study among Latino males (n=40; 35-55 years) residing in the Atlanta metro area.
4. To explore preliminary signals for improvements in objectively-measured PA, dietary behaviors, physical fitness and body composition outcomes with our soccer-based DPP adaptation.

Jiten Chhabra, MBBS, MS
Research Scientist
Georgia Institute of Technology

A Field Trial of the Guide To Goals Application – a Care Coordination Tool to Translate ADA Clinical Standards of Care for Children with T2D into Practice

Project Summary:

The purpose of the proposed research is to evaluate the Guide to Goals (GTG) application, a custom-made web-based care coordination tool that provides children, caretakers, and providers, easy access to individualized information to inform patient-specific decisions regarding the management of childhood type 2 diabetes (T2D).

Project Goals:

1. Conduct Usability Testing of the GTG Application Since its development, the GTG app has not been systematically tested with all potential users. We will conduct formal usability testing of the app with 5users from each of the stakeholder categories (patients/caretakers, front-desk staff, nurses, certified diabetes educators and physicians). Testing will be conducted in a task-based manner where researchers will observe and interview the participants as they use the existing app to complete critical tasks like completing survey instruments and using the application to set goals. Resulting data will be compiled into specifications and shared with the development team to make modifications before commencing the user trial described in Aim #2.
2. Conduct a User Trial of the modified GTG Application We will conduct a user trial of the modified GTG application in an office setting at the Pediatric Endocrinology Department at Children’s Healthcare of Atlanta. The primary goal of the trial will be to evaluate if the adherent use of the GTG application by the patient and their care team results in an improvement in mean blood glucose. Secondary goals are to evaluate goal setting (frequency and quality) when compared to the current standard of care.

Ambar Kulshreshtha, MD, PhD, FAHA
Assistant Professor, Department of Family and Preventive Medicine
Emory University School of Medicine

Culturally Tailored – Cognitive Behavioral Intervention (CT-CB) for African American patients with Type-2 Diabetes A Pilot Study

Project Summary:

African Americans are approximately twice as likely to be diagnosed with diabetes and to experience gaps in diabetes care compared to Whites. Lower health literacy and socio-economic, language, and communication barriers are all associated with disparities in diabetes care. There is evidence that behavioral interventions can help address some of these barriers. However, there is scarce data regarding the effectiveness of adapted and tailored cognitive-behavior change interventions and delivery in innovative low-cost formats for vulnerable population subgroups.

The aim of the study is to test the feasibility and acceptability of a Culturally-Tailored, Cognitive Behavioral intervention (CT-CB) program (in group-based and phone-based formats), compared with usual care, and to determine if the intervention can improve diabetes self-management among African American patients. We also seek to determine whether a particular subset of patients is more likely to experience improvements.

Project Goals:

1. Assess and compare the feasibility, and acceptability of a Culturally-Tailored Cognitive Behavior intervention (CT-CB) for African American patients with uncontrolled diabetes (HbA1C >8%) in a 3-armed randomized controlled study (group-based CT-CB, phone-based CT-CB, and usual care). Feasibility is operationalized as patients enrolled as per target with a loss to follow up of no more than 20%. Acceptability is measured as session attendance (>70%) and qualitative feedback from participants.
2. Assess the effects of the CT-CB intervention (group-based and phone-based) on secondary outcomes such as diabetes control (HbA1C), and glucose self-monitoring (self-reported) compared with usual individual-based care in clinic.
3. Determine if the CT-CB program as tested in the two formats (group-based and phone-based) can improve diabetes outcomes and self-management among particular subsets of African American patients (with poorly controlled vs. mild uncontrolled diabetes; < 65 years vs. > 65 years; men vs women) compared with usual individual-based care in clinic.

Ashely Coombe-Helvig, PhD, RN, CNE
Assistant Professor, Nell Hodgson Woodruff School of Nursing
Emory University

A Sleep Intervention (SLEEP-Extend) for Young Adults At-Risk for Type 2 Diabetes

Project Summary:

Recent guidelines published by the National Sleep Foundation (NSF) suggest that young adults 18-25 years obtain 7-9 hours of sleep per night (NSF, NSF recommends new sleep times). Insufficient sleep and sleep deprivation is becoming more common in young adults (Petrov et al., 2014), and just under 1/3 of the individuals aged 18-24 years of age surveyed in 2009 complained of insufficient sleep ≥14 days in the past 30 days (Liu, Y. et al, 2013). A Gallop poll reported even more alarming numbers with 49% of the individuals polled between the ages of 18 and 29 reporting less than 6 hours of sleep/night (Jones, 2013). This finding is particularly concerning since short sleep duration was identified as an independent risk factor for the development of IR and T2DM (Reutrakul & Van Cauter, 2014). Our goal is to delay the onset of T2DM; thus, we are intervening with young adults –– an age where primary prevention efforts may be most effective.

Project Goals:

The specific aims for this pilot study of young adults ages 18-25 years old with IR (based on established cut points), short sleep (< 6 hours/night) and obesity (BMI ≥ 30) are to:

1. Evaluate the feasibility and acceptability of a tailored intervention to extend sleep duration (SLEEP-Extend)
2. Determine changes in sleep duration after SLEEP-Extend intervention
3. Evaluate IR levels before and after SLEEP-Extend
4. Explore the potential mediators of diet and physical activity between sleep duration and IR. Our hypotheses are: H1: Individuals undergoing SLEEP-Extend will have longer sleep duration and increased knowledge of sleep hygiene post-intervention compared to controls and, H2: A greater proportion of the SLEEP-Extend group will have a decrease in IR compared to the control group.

Jonathan Colasanti, MD
Associate Professor, Department of Medicine
Emory University School of Medicine

Intersection of Chronic Diseases: Understanding the Care Continuum for Patients Co-affected by HIV and Diabetes

Project Summary:

The HIV and diabetes care continua have been explored separately, largely relying on cross-sectional approaches. In order to understand continuous retention and maintenance of treatment goals achievement among people dually affected by HIV and diabetes, a longitudinal approach is needed.20 We propose to assess the HIV-diabetes care continuum using a cross-sectional and a longitudinal approach. We will use data from the U.S. Military HIV Natural History Study (NHS),21 an ongoing prospective multicenter observational study exploring the natural history of HIV among U.S. military beneficiaries (see letter of support). Our cross- sectional preliminary work shows important gaps in diabetes care for HIV-positive women with diverse healthcare coverage; the present analysis will allow us to explore, longitudinally, how healthcare insurance impacts diabetes care and control. This is central to healthcare coverage policy for our armed service men and women living with HIV and diabetes.

Project Goals:

1. To estimate the proportion of patients that are aware of their diabetes diagnosis, linked to care, retained in care, and achieving diabetes care goals at three cross-sectional time points. This will be a cross-sectional analysis among HIV-positive patients from NHS (~6,000). From this pool, we will identify patients with diabetes and estimate what proportion of them are aware of their diabetes diagnosis, linked to care and retained in care at three cross-sectional time points (years 2000-04, 2005-09, 2010-15). We will also estimate what proportion of patients are achieving diabetes care goals (HbA1c <7%, blood pressure [systolic blood pressure <140 mmHg] and diastolic blood pressure <90 mmHg], cholesterol [LDL <100 mg/dL], and avoiding tobacco use). We will then compare diabetes care goal achievement between virally suppressed (viral load of HIV RNA of <200 copies/mL7) and not virally suppressed patients.
2. To explore continuous retention in care and achievement of HIV and diabetes treatment goals among people dually affected by HIV and diabetes. This will be a longitudinal analysis among HIV-positive patients with diabetes from NHS identified in Aim 1. The starting point for the analysis will be incident diabetes and from that point, patients will be followed to 1, 3, and 5 years after their diagnosis. We will examine differences in the proportion for patients retained and/or virally suppressed, and/or achieving diabetes treatment goals between time points. We will then compare the relative risks for achievement of multiple diabetes treatment goals between virally suppressed and not virally suppressed patients in all longitudinal time points using log-binomial models with a generalized estimating equation approach. Finally, since retirement from active duty can result in healthcare system change, we will explore how this transition impact retention in care and HIV-diabetes treatment outcomes using an interrupted time series approach.

Mary Beth Weber, PhD, MPH
Assistant Professor, Hubert Department of Global Health, Department of Epidemiology
Rollins School of Public Health, Emory University

Lauren Wilcox, PhD
Associate Professor, College of Computing
Georgia Institute of Technology

Mobile Health Based Type 2 Diabetes Management in South Asian Immigrants

Project Summary:

A community-based survey of South Asian (SA) immigrants living in Atlanta reported a T2D prevalence of 18.3%,25 markedly higher than the rates seen in the general US population (14.3%). In this proposal we aim to address the unmet need around management if T2D in the rapidly growing SA immigrant population with LEP in metro Atlanta. The Clarkston Community Health Center (CCHC) is a non-profit, free health clinic providing medical care for limited–English proficiency (LEP) refugee and immigrant populations in the Atlanta metropolitan area. CCHC provides primary care services to uninsured SA patients and will be the primary recruitment site. We will develop Hindi, Gujarati and Urdu (the three most commonly spoken languages in SA26) language-based mhealth educational materials and modules based on DSME/S strategies and community feedback and test their effectiveness in improving T2D control in a pilot study. This mHealth app will also provide patient management tools for the physicians treating these diabetes patients to maximize patient care and provider feedback. We hypothesize that this language- based culturally-sensitive mhealth app will eliminate the disparities and challenges in management of T2D among patients with LEP, improve patient engagement in self-management of their T2D and improve control of T2D.

Project Goals:

1. To develop an innovative, multilingual (Hindi, Gujarati, and Urdu), smartphone-based mhealth app with modules on T2D education and self-management that targets LEP immigrants of SA origin.
2. To examine the impact of the language-specific mhealth app on diabetic control (primary outcome: hemoglobin A1C) in a randomized study comparing the mhealth app with a control group that will receive printed educational materials in their language and conventional provider treatment during clinic visits. Our hypothesis is that, compared to control participants, SA individuals with diabetes randomized to the mhealth app study arm will reduce HbA1c levels by 1.45% more.
3. To conduct in-depth interviews with information-rich study participants and study physicians to describe the acceptability, usability and usefulness of the mhealth app. This study will develop and test the effectiveness, acceptability, and usability of materials for online T2D education in multiple languages spoken by SA immigrants. Not only will findings from this pilot trial guide the design of a future definitive clinical trial, but they may be used to guide the development of similar programs in other migrant or LEP populations.

Solveig Cunningham, MD, MSc
Associate Professor, Hubert Department of Global Health
Rollins School of Public Health, Emory University

Adoption of New Lifestyles among Refugee Families: Health Implications of Integration

Project Summary:

The overarching aim of this project is to advance knowledge about the processes by which newly arrived parents and youths experience increases in obesity risks with residence in the U.S., a process we call bodyweight acculturation. We aim to develop, in partnership with a partner NGO working with this population, recommendations for promoting healthy behavior on arrival with the goal of preventing health disparities before they emerge.

Project Goals:

1. Quantify through direct anthropometric measurements changes in body weight in newly-arrived refugee mother and adolescent pairs during the first year since arrival to the U.S.
2. Identify via quantitative survey instruments the timing of changes in diet and activity levels during the first year since arrival to the U.S. among newly-arrived refugee mother and adolescent pairs.
3. Evaluate via qualitative data collection newly-arrived refugee adolescents’ and mothers’ interaction with and interpretation of their new food environment.
4. Propose evidence-based recommendations for timing and methods of promoting healthy acculturation among newly arrived youths and their families.