Kimberly A. Dukes
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Kimberly A. Dukes has a total of 61 co-authors. Here are the top co-authors by number of shared publications:
Abhi Jain
Department of Biostatistics, Boston University School of Public Health, Boston, USA
3 shared publicationsMichael LaValley
Department of Biostatistics, Boston University School of Public Health, Boston, USA
3 shared publicationsMichael Rickles
Sharecare, Inc, Atlanta, USA
3 shared publicationsTara Tripp
Department of Biostatistics, Boston University School of Public Health, Boston, USA
3 shared publicationsBiqi Wang
Department of Medicine, UMass Chan Medical School, Worcester, MA, USA
2 shared publicationsBrandeus Davis
Sharecare, Inc, Atlanta, USA
2 shared publications
Latest Publications
- Open Access
- 22 July 2026
- Published by Springer Nature in Population Health Metrics
- Vol. 24 (1)
Characterizing U.S. community well-being indices: a methodological synthesis and measurement systems perspective
- 22 July 2026
- Published by Springer Nature in Population Health Metrics
- Vol. 24 (1)
Open AccessCharacterizing U.S. community well-being indices: a methodological synthesis and measurement systems perspective
- Open Access
- 25 December 2025
- Published by Springer Nature in Addiction Science & Clinical Practice
- Vol. 21 (1)
Pre-exposure prophylaxis (PrEP) and medications for opioid use disorder for persons who inject drugs: the CHORUS + randomized controlled trial study protocol
Background: Human immunodeficiency virus (HIV) cases among people who inject drugs increased during the US overdose crisis. Although HIV pre-exposure prophylaxis (PrEP) decreases HIV acquisition, and medications for opioid use disorder (MOUD) reduce overdose deaths, uptake remains suboptimal. The CHORUS + study will test the efficacy of a comprehensive peer recovery coaching intervention to increase PrEP and MOUD initiation and adherence. Methods: This two-arm RCT will enroll 284 people who inject opioids (PWIO) and are negative for HIV from two sites. Participants randomized to the CHORUS + intervention will receive a study smartphone, motivational interviewing sessions, assistance connecting to PrEP and/or MOUD, and support to access resources addressing social needs such as employment and housing. We will also incorporate adapted ‘Life-Steps for PrEP’ modules to increase adherence. The control arm will receive information on local organizations with access to PrEP and MOUD. Participants will complete assessments at baseline, 1-, 3-, 6-, and 12-month visits. The primary outcome is adherence to HIV PrEP, assessed by tenofovir-diphosphate drug levels at 6-months post enrollment. Secondary outcomes include PrEP adherence assessed at 3- and 12-months, measured by drug levels (3-months), prescription refills, and self-report; and MOUD receipt at 3-, 6-, and 12-months, measured by prescription refills, self-report, and urine toxicology. The primary analysis will employ intent-to-treat logistic regression to assess differences in adherence between treatment arms, adjusting for stratification factors including site, race and sex assigned at birth. We will analyze secondary outcomes using similar methods. We will use multilevel growth curve modeling to evaluate changes in adherence over time by treatment group, incorporating random intercepts and slopes to account for individual trajectories. We will use exploratory multilevel structural equation modeling to assess mediators including HIV risk perception and PrEP/MOUD knowledge to understand pathways that may influence adherence. Discussion: The CHORUS + intervention integrates a novel, theory-based, peer-delivered, smartphone-supported approach to address HIV prevention and opioid use disorder, while tackling social and structural barriers to care. Findings will inform strategies for linking PWIO to rapid HIV prevention and substance use treatment. Trial registration: ClinicalTrials.gov number: NCT05769218.
- 25 December 2025
- Published by Springer Nature in Addiction Science & Clinical Practice
- Vol. 21 (1)
Open AccessPre-exposure prophylaxis (PrEP) and medications for opioid use disorder for persons who inject drugs: the CHORUS + randomized controlled trial study protocol
Background: Human immunodeficiency virus (HIV) cases among people who inject drugs increased during the US overdose crisis. Although HIV pre-exposure prophylaxis (PrEP) decreases HIV acquisition, and medications for opioid use disorder (MOUD) reduce overdose deaths, uptake remains suboptimal. The CHORUS + study will test the efficacy of a comprehensive peer recovery coaching intervention to increase PrEP and MOUD initiation and adherence. Methods: This two-arm RCT will enroll 284 people who inject opioids (PWIO) and are negative for HIV from two sites. Participants randomized to the CHORUS + intervention will receive a study smartphone, motivational interviewing sessions, assistance connecting to PrEP and/or MOUD, and support to access resources addressing social needs such as employment and housing. We will also incorporate adapted ‘Life-Steps for PrEP’ modules to increase adherence. The control arm will receive information on local organizations with access to PrEP and MOUD. Participants will complete assessments at baseline, 1-, 3-, 6-, and 12-month visits. The primary outcome is adherence to HIV PrEP, assessed by tenofovir-diphosphate drug levels at 6-months post enrollment. Secondary outcomes include PrEP adherence assessed at 3- and 12-months, measured by drug levels (3-months), prescription refills, and self-report; and MOUD receipt at 3-, 6-, and 12-months, measured by prescription refills, self-report, and urine toxicology. The primary analysis will employ intent-to-treat logistic regression to assess differences in adherence between treatment arms, adjusting for stratification factors including site, race and sex assigned at birth. We will analyze secondary outcomes using similar methods. We will use multilevel growth curve modeling to evaluate changes in adherence over time by treatment group, incorporating random intercepts and slopes to account for individual trajectories. We will use exploratory multilevel structural equation modeling to assess mediators including HIV risk perception and PrEP/MOUD knowledge to understand pathways that may influence adherence. Discussion: The CHORUS + intervention integrates a novel, theory-based, peer-delivered, smartphone-supported approach to address HIV prevention and opioid use disorder, while tackling social and structural barriers to care. Findings will inform strategies for linking PWIO to rapid HIV prevention and substance use treatment. Trial registration: ClinicalTrials.gov number: NCT05769218.
- Open Access
- 21 November 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 41272
A national growth mixture modeling analysis of county-level COVID-19 incidence rate trajectories and health inequities during three successive pandemic waves in 2020
We applied Growth Mixture Modeling (GMM) to characterize county-level COVID-19 incidence rate (IR) trajectories across three distinct waves in the United States from March 15 to November 2, 2020. GMM enabled the identification of latent subpopulations with shared temporal patterns of disease spread, offering a flexible analytic framework for uncovering both known and evolving disparities. Across the three periods, up to five trajectory groups were identified, revealing substantial geographic and temporal heterogeneity. To support interpretation and reduce multiple comparisons, we developed and validated a 17-item Social Determinants of Health (SDOH) index representing county-level economic and resource access factors. Higher-incidence rate trajectories consistently aligned with lower SDOH scores and greater proportions of Black or African American residents and younger populations. These disparities shifted over time, with patterns of high-incidence rates emerging in resource-limited counties across the South and Midwest in later waves. By further assessing well-documented inequities and expanding understanding of their dynamic expression across space and time, this study demonstrates the utility of GMM for public health surveillance, planning, and equitable response strategies in future outbreaks.
- 21 November 2025
- Published by Springer Nature in Scientific Reports
- Vol. 15 (1) , 41272
Open AccessA national growth mixture modeling analysis of county-level COVID-19 incidence rate trajectories and health inequities during three successive pandemic waves in 2020
We applied Growth Mixture Modeling (GMM) to characterize county-level COVID-19 incidence rate (IR) trajectories across three distinct waves in the United States from March 15 to November 2, 2020. GMM enabled the identification of latent subpopulations with shared temporal patterns of disease spread, offering a flexible analytic framework for uncovering both known and evolving disparities. Across the three periods, up to five trajectory groups were identified, revealing substantial geographic and temporal heterogeneity. To support interpretation and reduce multiple comparisons, we developed and validated a 17-item Social Determinants of Health (SDOH) index representing county-level economic and resource access factors. Higher-incidence rate trajectories consistently aligned with lower SDOH scores and greater proportions of Black or African American residents and younger populations. These disparities shifted over time, with patterns of high-incidence rates emerging in resource-limited counties across the South and Midwest in later waves. By further assessing well-documented inequities and expanding understanding of their dynamic expression across space and time, this study demonstrates the utility of GMM for public health surveillance, planning, and equitable response strategies in future outbreaks.
- Open Access
- 22 April 2024
- Published by Springer Nature in Scientific Reports
- Vol. 14 (1) , 1-12
Modeling health and well-being measures using ZIP code spatial neighborhood patterns
Individual-level assessment of health and well-being permits analysis of community well-being and health risk evaluations across several dimensions of health. It also enables comparison and rankings of reported health and well-being for large geographical areas such as states, metropolitan areas, and counties. However, there is large variation in reported well-being within such large spatial units underscoring the importance of analyzing well-being at more granular levels, such as ZIP codes. In this paper, we address this problem by modeling well-being data to generate ZIP code tabulation area (ZCTA)-level rankings through spatially informed statistical modeling. We build regression models for individual-level overall well-being index and scores from five subscales (Physical, Financial, Social, Community, Purpose) using individual-level demographic characteristics as predictors while including a ZCTA-level spatial effect. The ZCTA neighborhood information is incorporated by using a graph Laplacian matrix; this enables estimation of the effect of a ZCTA on well-being using individual-level data from that ZCTA as well as by borrowing information from neighboring ZCTAs. We deploy our model on well-being data for the U.S. states of Massachusetts and Georgia. We find that our model can capture the effects of demographic features while also offering spatial effect estimates for all ZCTAs, including ones with no observations, under certain conditions. These spatial effect estimates provide community health and well-being rankings of ZCTAs, and our method can be deployed more generally to model other outcomes that are spatially dependent as well as data from other states or groups of states.
- 22 April 2024
- Published by Springer Nature in Scientific Reports
- Vol. 14 (1) , 1-12
Open AccessModeling health and well-being measures using ZIP code spatial neighborhood patterns
Individual-level assessment of health and well-being permits analysis of community well-being and health risk evaluations across several dimensions of health. It also enables comparison and rankings of reported health and well-being for large geographical areas such as states, metropolitan areas, and counties. However, there is large variation in reported well-being within such large spatial units underscoring the importance of analyzing well-being at more granular levels, such as ZIP codes. In this paper, we address this problem by modeling well-being data to generate ZIP code tabulation area (ZCTA)-level rankings through spatially informed statistical modeling. We build regression models for individual-level overall well-being index and scores from five subscales (Physical, Financial, Social, Community, Purpose) using individual-level demographic characteristics as predictors while including a ZCTA-level spatial effect. The ZCTA neighborhood information is incorporated by using a graph Laplacian matrix; this enables estimation of the effect of a ZCTA on well-being using individual-level data from that ZCTA as well as by borrowing information from neighboring ZCTAs. We deploy our model on well-being data for the U.S. states of Massachusetts and Georgia. We find that our model can capture the effects of demographic features while also offering spatial effect estimates for all ZCTAs, including ones with no observations, under certain conditions. These spatial effect estimates provide community health and well-being rankings of ZCTAs, and our method can be deployed more generally to model other outcomes that are spatially dependent as well as data from other states or groups of states.
- Open Access
- 23 August 2021
- Published by American Medical Association (AMA) in JAMA Network Open
- Vol. 4 (8) , e2121726
Association of Prenatal Exposure to Maternal Drinking and Smoking With the Risk of Stillbirth
Stillbirth is a worldwide problem, with a reported overall rate in 2015 of 18.4 per 1000 births.1 In 2013 in the US, there were 23 596 stillbirths, with a rate of 5.96 per 1000 births.1 Identification of modifiable risk factors may reduce the number of these deaths.2 A strong association between maternal smoking during pregnancy and stillbirth has been reported.3-5 Although associations have been found for maternal drinking during pregnancy and abortion, being small for gestational age, and stillbirth,6-8 the association with stillbirth may be limited to binge drinking.9 Previous studies9,10 that assessed the association between stillbirth and maternal drinking or smoking have often been retrospective and have inadequately classified quantity, frequency, and timing of exposures. In addition, many of the stillbirth classification systems used do not sufficiently capture underlying disease processes in the exposed fetus or placenta and often use inconsistent or incomplete pathology protocols.
- 23 August 2021
- Published by American Medical Association (AMA) in JAMA Network Open
- Vol. 4 (8) , e2121726
Open AccessAssociation of Prenatal Exposure to Maternal Drinking and Smoking With the Risk of Stillbirth
Stillbirth is a worldwide problem, with a reported overall rate in 2015 of 18.4 per 1000 births.1 In 2013 in the US, there were 23 596 stillbirths, with a rate of 5.96 per 1000 births.1 Identification of modifiable risk factors may reduce the number of these deaths.2 A strong association between maternal smoking during pregnancy and stillbirth has been reported.3-5 Although associations have been found for maternal drinking during pregnancy and abortion, being small for gestational age, and stillbirth,6-8 the association with stillbirth may be limited to binge drinking.9 Previous studies9,10 that assessed the association between stillbirth and maternal drinking or smoking have often been retrospective and have inadequately classified quantity, frequency, and timing of exposures. In addition, many of the stillbirth classification systems used do not sufficiently capture underlying disease processes in the exposed fetus or placenta and often use inconsistent or incomplete pathology protocols.