Hayoung Jeong

Graduate Trainee, Duke University

9 active projects

RT- Social Determinants of Health and Average Steps

These scientific inquiries aim to elucidate the intricate connections between social factors and individuals' levels of physical activity. The relevance of these questions lies in their potential to uncover health disparities among diverse populations based on various social determinants, such…

Scientific Questions Being Studied

These scientific inquiries aim to elucidate the intricate connections between social factors and individuals' levels of physical activity. The relevance of these questions lies in their potential to uncover health disparities among diverse populations based on various social determinants, such as income, education, and neighborhood environment. Understanding these disparities is vital for devising targeted interventions that can effectively reduce health inequities.

Project Purpose(s)

  • Methods Development

Scientific Approaches

Fitbit data with social determinants data will be merged and integrated, ensuring that each individual's physical activity records are linked with their corresponding social determinants information. We will conduct an initial descriptive analysis to understand the basic statistics and patterns in the data. Explore We will also use statistical methods to determine the correlations between social determinants and average step counts, and regression analysis will be performed to quantify the impact of social determinants on average step counts while controlling for potential confounding variables.

Anticipated Findings

Identification of Significant Correlations: The study may reveal statistically significant correlations between specific social determinants (such as income, education, neighborhood environment, or access to recreational facilities) and average daily step counts. These correlations may vary across demographic groups.
Impact on Health Disparities: It is likely that the study will find evidence of health disparities, with certain groups facing lower physical activity levels due to disadvantaged social determinants. This could include individuals from lower-income neighborhoods or with limited access to healthy food options.
Mediating and Moderating Factors: The research might identify factors that mediate or moderate the relationship between social determinants and physical activity. For example, access to parks may mediate the impact of neighborhood environment on step counts, or age may moderate the relationship between education and physical activity.

Demographic Categories of Interest

This study will not center on underrepresented populations.

Data Set Used

Controlled Tier

Research Team

Owner:

Social Determinants of Health and Average Steps

These scientific inquiries aim to elucidate the intricate connections between social factors and individuals' levels of physical activity. The relevance of these questions lies in their potential to uncover health disparities among diverse populations based on various social determinants, such…

Scientific Questions Being Studied

These scientific inquiries aim to elucidate the intricate connections between social factors and individuals' levels of physical activity. The relevance of these questions lies in their potential to uncover health disparities among diverse populations based on various social determinants, such as income, education, and neighborhood environment. Understanding these disparities is vital for devising targeted interventions that can effectively reduce health inequities.

Project Purpose(s)

  • Methods Development

Scientific Approaches

Fitbit data with social determinants data will be merged and integrated, ensuring that each individual's physical activity records are linked with their corresponding social determinants information. We will conduct an initial descriptive analysis to understand the basic statistics and patterns in the data. Explore We will also use statistical methods to determine the correlations between social determinants and average step counts, and regression analysis will be performed to quantify the impact of social determinants on average step counts while controlling for potential confounding variables.

Anticipated Findings

Identification of Significant Correlations: The study may reveal statistically significant correlations between specific social determinants (such as income, education, neighborhood environment, or access to recreational facilities) and average daily step counts. These correlations may vary across demographic groups.
Impact on Health Disparities: It is likely that the study will find evidence of health disparities, with certain groups facing lower physical activity levels due to disadvantaged social determinants. This could include individuals from lower-income neighborhoods or with limited access to healthy food options.
Mediating and Moderating Factors: The research might identify factors that mediate or moderate the relationship between social determinants and physical activity. For example, access to parks may mediate the impact of neighborhood environment on step counts, or age may moderate the relationship between education and physical activity.

Demographic Categories of Interest

This study will not center on underrepresented populations.

Data Set Used

Controlled Tier

Research Team

Owner:

Missing Analysis 2

A key objective is to develop data-driven methods which to determine the most effective adherence thresholds. This is crucial for ensuring that the data under analysis meet a certain quality standard and are relevant for the intended research purposes. Another…

Scientific Questions Being Studied

A key objective is to develop data-driven methods which to determine the most effective adherence thresholds. This is crucial for ensuring that the data under analysis meet a certain quality standard and are relevant for the intended research purposes. Another significant aim is to thoroughly define and analyze the patterns of missingness in the data, both on an intraday and daily basis. This involves applying clustering techniques to discern these patterns among individuals and across the population. his approach will not only help in understanding the nuances of data missingness but also in identifying potential systemic biases or errors in data collection and processing.

Project Purpose(s)

  • Methods Development

Scientific Approaches

Filter data based on HR filtering (used in AoU) or other definitions in the literature to ensure data quality.
Visualization: Developing figures, such as (looking like) elbow diagrams, to identify optimal adherence thresholds.

Defining and analyzing missingness patterns (intraday & daily)
Applying clustering to understand the missingness pattern both at the individual level and across the population (what Peter did).
Making associations (correlation) with outcome variables or covariates

Anticipated Findings

Firstly, we expect to identify optimal data filtering criteria that align with HR filtering standards or other reputable definitions, which will set a benchmark for data quality in subsequent analyses. Through the development of visualizations like elbow diagrams, the study aims to pinpoint precise adherence thresholds that ensure data reliability, offering a methodological advancement in data preprocessing for health research.

In analyzing missingness patterns, both intraday and daily, we anticipate uncovering significant insights into the nature and extent of data gaps. By applying clustering techniques, we aim to categorize individuals and populations based on their missingness profiles, potentially revealing underlying reasons for data absence, such as demographic factors, study design issues, or data collection methodologies. This could lead to recommendations for improving data collection processes and minimizing missing data in future research.

Demographic Categories of Interest

This study will not center on underrepresented populations.

Data Set Used

Controlled Tier

Research Team

Owner:

Collaborators:

  • Shun Sakai - Undergraduate Student, Duke University
  • Yuyou Wu - Graduate Trainee, Duke University
  • Lori Liu - Graduate Trainee, Duke University
  • Jiamu Yang - Graduate Trainee, Duke University
  • Harrison Kane - Undergraduate Student, Duke University

Archived: Missing Analysis

A key objective is to develop data-driven methods which to determine the most effective adherence thresholds. This is crucial for ensuring that the data under analysis meet a certain quality standard and are relevant for the intended research purposes. Another…

Scientific Questions Being Studied

A key objective is to develop data-driven methods which to determine the most effective adherence thresholds. This is crucial for ensuring that the data under analysis meet a certain quality standard and are relevant for the intended research purposes. Another significant aim is to thoroughly define and analyze the patterns of missingness in the data, both on an intraday and daily basis. This involves applying clustering techniques to discern these patterns among individuals and across the population. his approach will not only help in understanding the nuances of data missingness but also in identifying potential systemic biases or errors in data collection and processing.

Project Purpose(s)

  • Methods Development

Scientific Approaches

Filter data based on HR filtering (used in AoU) or other definitions in the literature to ensure data quality.
Visualization: Developing figures, such as (looking like) elbow diagrams, to identify optimal adherence thresholds.

Defining and analyzing missingness patterns (intraday & daily)
Applying clustering to understand the missingness pattern both at the individual level and across the population (what Peter did).
Making associations (correlation) with outcome variables or covariates

Anticipated Findings

Firstly, we expect to identify optimal data filtering criteria that align with HR filtering standards or other reputable definitions, which will set a benchmark for data quality in subsequent analyses. Through the development of visualizations like elbow diagrams, the study aims to pinpoint precise adherence thresholds that ensure data reliability, offering a methodological advancement in data preprocessing for health research.

In analyzing missingness patterns, both intraday and daily, we anticipate uncovering significant insights into the nature and extent of data gaps. By applying clustering techniques, we aim to categorize individuals and populations based on their missingness profiles, potentially revealing underlying reasons for data absence, such as demographic factors, study design issues, or data collection methodologies. This could lead to recommendations for improving data collection processes and minimizing missing data in future research.

Demographic Categories of Interest

This study will not center on underrepresented populations.

Data Set Used

Controlled Tier

Research Team

Owner:

Collaborators:

  • Shun Sakai - Undergraduate Student, Duke University
  • Yuyou Wu - Graduate Trainee, Duke University
  • Lori Liu - Graduate Trainee, Duke University
  • Jiamu Yang - Graduate Trainee, Duke University
  • Harrison Kane - Undergraduate Student, Duke University

Duplicate of Demo Project: State-level Activity Inequality [Published Work]

How is physical activity distributed within states in the US? Analysis of such activity distributions and inequality can reveal important relationships between physical activity disparities, health outcomes, and modifiable factors, as Althoff et al. studied in their paper, "Large-scale physical…

Scientific Questions Being Studied

How is physical activity distributed within states in the US? Analysis of such activity distributions and inequality can reveal important relationships between physical activity disparities, health outcomes, and modifiable factors, as Althoff et al. studied in their paper, "Large-scale physical activity data reveal worldwide activity inequality" (2017).

Project Purpose(s)

  • Educational

Scientific Approaches

The cohort will consist of Fitbit users in the US, with analysis being subdivided to the state level. Various graphs will be utilized to help visualize the low- and high-activity trends across states. Well-defined measures such as the Gini coefficient will be used to aid in the analysis of activity inequality.

Anticipated Findings

The study aims to find relationships between activity inequality and health outcomes, such as obesity levels. With the growing accessibility of fitness trackers and activity sensors built into personal devices, this study hopes to leverage the volume of available data and potentially inform measures to improve population activity and health.

Demographic Categories of Interest

This study will not center on underrepresented populations.

Data Set Used

Controlled Tier

Research Team

Owner:

Collaborators:

  • Srikar Katta - Graduate Trainee, Duke University

DST_Sleep_Analysis_2.0

NA

Scientific Questions Being Studied

NA

Project Purpose(s)

  • Population Health

Scientific Approaches

NA

Anticipated Findings

NA

Demographic Categories of Interest

This study will not center on underrepresented populations.

Data Set Used

Controlled Tier

Research Team

Owner:

Demo Project: State-level Activity Inequality [Published Work]

How is physical activity distributed within states in the US? Analysis of such activity distributions and inequality can reveal important relationships between physical activity disparities, health outcomes, and modifiable factors, as Althoff et al. studied in their paper, "Large-scale physical…

Scientific Questions Being Studied

How is physical activity distributed within states in the US? Analysis of such activity distributions and inequality can reveal important relationships between physical activity disparities, health outcomes, and modifiable factors, as Althoff et al. studied in their paper, "Large-scale physical activity data reveal worldwide activity inequality" (2017).

Project Purpose(s)

  • Educational

Scientific Approaches

The cohort will consist of Fitbit users in the US, with analysis being subdivided to the state level. Various graphs will be utilized to help visualize the low- and high-activity trends across states. Well-defined measures such as the Gini coefficient will be used to aid in the analysis of activity inequality.

Anticipated Findings

The study aims to find relationships between activity inequality and health outcomes, such as obesity levels. With the growing accessibility of fitness trackers and activity sensors built into personal devices, this study hopes to leverage the volume of available data and potentially inform measures to improve population activity and health.

Demographic Categories of Interest

This study will not center on underrepresented populations.

Data Set Used

Controlled Tier

Research Team

Owner:

DST_Sleep_Analysis

NA

Scientific Questions Being Studied

NA

Project Purpose(s)

  • Population Health

Scientific Approaches

NA

Anticipated Findings

NA

Demographic Categories of Interest

This study will not center on underrepresented populations.

Data Set Used

Controlled Tier

Research Team

Owner:

Duplicate of Wearables and The Human Phenome (Published Work)

Our primary goal is to understand the relation between activity levels with the development and progression of human disease. Higher physical activity is associated with lower prevalence and better outcomes in virtually every human disease. These analyses will generate hypotheses…

Scientific Questions Being Studied

Our primary goal is to understand the relation between activity levels with the development and progression of human disease. Higher physical activity is associated with lower prevalence and better outcomes in virtually every human disease. These analyses will generate hypotheses guiding clinical and research interventions focused on activity to reduce morbidity and mortality in patients seeking care.

This workspace is replication workspace for Wearables and The Human Phenome project. We replicated the workspace to provide a clean and reduced version of code that was used to generate the findings, which were published in Nature Medicine (https://www.nature.com/articles/s41591-022-02012-w).

Project Purpose(s)

  • Population Health
  • Social / Behavioral

Scientific Approaches

We will examine the relationship between daily activity (steps, activity intensity) over time and the prevalence and progression of coded human diseases. We will use the Fitbit data, EHR-curated diagnoses, laboratory values, and survey results.

Anticipated Findings

We expect to find that lower levels of activity are associated with a higher prevalence and more rapid progression of chronic diseases. These data will provide the rationale to link wearables data with electronic health records nationwide as a window into behavioral activity choice as a modifiable risk factor for chronic diseases. We may find substantial variation in activity and disease prevalence/severity by socioeconomic status, which would motivate studies/interventions to reduce these health disparities.

Demographic Categories of Interest

  • Race / Ethnicity
  • Geography
  • Access to Care
  • Education Level
  • Income Level

Data Set Used

Registered Tier

Research Team

Owner:

1 - 9 of 9
<
>
Request a Review of this Research Project

You can request that the All of Us Resource Access Board (RAB) review a research purpose description if you have concerns that this research project may stigmatize All of Us participants or violate the Data User Code of Conduct in some other way. To request a review, you must fill in a form, which you can access by selecting ‘request a review’ below.