Thomas Thornhill Iv
Research Subjects
Thomas Thornhill Iv has a total of 34 co-authors. Here are the top co-authors by number of shared publications:
- Dane Anthony Morey
The Ohio State University, Columbus, OH, USA
2 shared publications - Dante Della Vella
Ohio State Univ, Coll Engn, Dept Integrated Syst Engn, Columbus, OH USA
2 shared publications - Daria Faulkner
Ohio State Univ, Coll Publ Hlth, Columbus, OH USA
2 shared publications - David Kline
Department of Biostatistics and Data Science, Wake Forest University School of Medicine, Winston-Salem, NC
2 shared publications - Gregg S. Gonsalves
Department of Epidemiology of Microbial Diseases, Yale School of Public Health, New Haven, CT, USA
2 shared publications - Judson A. Brewer
Department of Psychiatry, Yale University School of Medicine, New Haven, CT 06511, USA
2 shared publications
Latest Publications
- Open Access
- 15 August 2023
- Published by JMIR Publications Inc. in JMIR Public Health and Surveillance
- Vol. 9 (1) , e39754
Using Bandit Algorithms to Maximize SARS-CoV-2 Case-Finding: Evaluation and Feasibility Study
Background: The Flexible Adaptive Algorithmic Surveillance Testing (FAAST) program represents an innovative approach for improving the detection of new cases of infectious disease; it is deployed here to screen and diagnose SARS-CoV-2. With the advent of treatment for COVID-19, finding individuals infected with SARS-CoV-2 is an urgent clinical and public health priority. While these kinds of Bayesian search algorithms are used widely in other settings (eg, to find downed aircraft, in submarine recovery, and to aid in oil exploration), this is the first time that Bayesian adaptive approaches have been used for active disease surveillance in the field. Objective: This study’s objective was to evaluate a Bayesian search algorithm to target hotspots of SARS-CoV-2 transmission in the community with the goal of detecting the most cases over time across multiple locations in Columbus, Ohio, from August to October 2021. Methods: The algorithm used to direct pop-up SARS-CoV-2 testing for this project is based on Thompson sampling, in which the aim is to maximize the average number of new cases of SARS-CoV-2 diagnosed among a set of testing locations based on sampling from prior probability distributions for each testing site. An academic-governmental partnership between Yale University, The Ohio State University, Wake Forest University, the Ohio Department of Health, the Ohio National Guard, and the Columbus Metropolitan Libraries conducted a study of bandit algorithms to maximize the detection of new cases of SARS-CoV-2 in this Ohio city in 2021. The initiative established pop-up COVID-19 testing sites at 13 Columbus locations, including library branches, recreational and community centers, movie theaters, homeless shelters, family services centers, and community event sites. Our team conducted between 0 and 56 tests at the 16 testing events, with an overall average of 25.3 tests conducted per event and a moving average that increased over time. Small incentives—including gift cards and take-home rapid antigen tests—were offered to those who approached the pop-up sites to encourage their participation. Results: Over time, as expected, the Bayesian search algorithm directed testing efforts to locations with higher yields of new diagnoses. Surprisingly, the use of the algorithm also maximized the identification of cases among minority residents of underserved communities, particularly African Americans, with the pool of participants overrepresenting these people relative to the demographic profile of the local zip code in which testing sites were located. Conclusions: This study demonstrated that a pop-up testing strategy using a bandit algorithm can be feasibly deployed in an urban setting during a pandemic. It is the first real-world use of these kinds of algorithms for disease surveillance and represents a key step in evaluating the effectiveness of their use in maximizing the detection of undiagnosed cases of SARS-CoV-2 and other infections, such as HIV.
- 15 August 2023
- Published by JMIR Publications Inc. in JMIR Public Health and Surveillance
- Vol. 9 (1) , e39754
Open AccessUsing Bandit Algorithms to Maximize SARS-CoV-2 Case-Finding: Evaluation and Feasibility Study
Background: The Flexible Adaptive Algorithmic Surveillance Testing (FAAST) program represents an innovative approach for improving the detection of new cases of infectious disease; it is deployed here to screen and diagnose SARS-CoV-2. With the advent of treatment for COVID-19, finding individuals infected with SARS-CoV-2 is an urgent clinical and public health priority. While these kinds of Bayesian search algorithms are used widely in other settings (eg, to find downed aircraft, in submarine recovery, and to aid in oil exploration), this is the first time that Bayesian adaptive approaches have been used for active disease surveillance in the field. Objective: This study’s objective was to evaluate a Bayesian search algorithm to target hotspots of SARS-CoV-2 transmission in the community with the goal of detecting the most cases over time across multiple locations in Columbus, Ohio, from August to October 2021. Methods: The algorithm used to direct pop-up SARS-CoV-2 testing for this project is based on Thompson sampling, in which the aim is to maximize the average number of new cases of SARS-CoV-2 diagnosed among a set of testing locations based on sampling from prior probability distributions for each testing site. An academic-governmental partnership between Yale University, The Ohio State University, Wake Forest University, the Ohio Department of Health, the Ohio National Guard, and the Columbus Metropolitan Libraries conducted a study of bandit algorithms to maximize the detection of new cases of SARS-CoV-2 in this Ohio city in 2021. The initiative established pop-up COVID-19 testing sites at 13 Columbus locations, including library branches, recreational and community centers, movie theaters, homeless shelters, family services centers, and community event sites. Our team conducted between 0 and 56 tests at the 16 testing events, with an overall average of 25.3 tests conducted per event and a moving average that increased over time. Small incentives—including gift cards and take-home rapid antigen tests—were offered to those who approached the pop-up sites to encourage their participation. Results: Over time, as expected, the Bayesian search algorithm directed testing efforts to locations with higher yields of new diagnoses. Surprisingly, the use of the algorithm also maximized the identification of cases among minority residents of underserved communities, particularly African Americans, with the pool of participants overrepresenting these people relative to the demographic profile of the local zip code in which testing sites were located. Conclusions: This study demonstrated that a pop-up testing strategy using a bandit algorithm can be feasibly deployed in an urban setting during a pandemic. It is the first real-world use of these kinds of algorithms for disease surveillance and represents a key step in evaluating the effectiveness of their use in maximizing the detection of undiagnosed cases of SARS-CoV-2 and other infections, such as HIV.
- Open Access
- 27 June 2022
- Published by JMIR Publications Inc. in JMIR Preprints
Using Bandit Algorithms to Maximize SARS-CoV-2 Case-Finding: Evaluation and Feasibility Study (Preprint)
BACKGROUND The Flexible Adaptive Algorithmic Surveillance Testing (FAAST) program represents an innovative approach for improving the detection of new cases of infectious disease; it is deployed here to screen and diagnose SARS-CoV-2. With the advent of treatment for COVID-19, finding individuals infected with SARS-CoV-2 is an urgent clinical and public health priority. While these kinds of Bayesian search algorithms are used widely in other settings (eg, to find downed aircraft, in submarine recovery, and to aid in oil exploration), this is the first time that Bayesian adaptive approaches have been used for active disease surveillance in the field. OBJECTIVE This study’s objective was to evaluate a Bayesian search algorithm to target hotspots of SARS-CoV-2 transmission in the community with the goal of detecting the most cases over time across multiple locations in Columbus, Ohio, from August to October 2021. METHODS The algorithm used to direct pop-up SARS-CoV-2 testing for this project is based on Thompson sampling, in which the aim is to maximize the average number of new cases of SARS-CoV-2 diagnosed among a set of testing locations based on sampling from prior probability distributions for each testing site. An academic-governmental partnership between Yale University, The Ohio State University, Wake Forest University, the Ohio Department of Health, the Ohio National Guard, and the Columbus Metropolitan Libraries conducted a study of bandit algorithms to maximize the detection of new cases of SARS-CoV-2 in this Ohio city in 2021. The initiative established pop-up COVID-19 testing sites at 13 Columbus locations, including library branches, recreational and community centers, movie theaters, homeless shelters, family services centers, and community event sites. Our team conducted between 0 and 56 tests at the 16 testing events, with an overall average of 25.3 tests conducted per event and a moving average that increased over time. Small incentives—including gift cards and take-home rapid antigen tests—were offered to those who approached the pop-up sites to encourage their participation. RESULTS Over time, as expected, the Bayesian search algorithm directed testing efforts to locations with higher yields of new diagnoses. Surprisingly, the use of the algorithm also maximized the identification of cases among minority residents of underserved communities, particularly African Americans, with the pool of participants overrepresenting these people relative to the demographic profile of the local zip code in which testing sites were located. CONCLUSIONS This study demonstrated that a pop-up testing strategy using a bandit algorithm can be feasibly deployed in an urban setting during a pandemic. It is the first real-world use of these kinds of algorithms for disease surveillance and represents a key step in evaluating the effectiveness of their use in maximizing the detection of undiagnosed cases of SARS-CoV-2 and other infections, such as HIV.
- 27 June 2022
- Published by JMIR Publications Inc. in JMIR Preprints
Open AccessUsing Bandit Algorithms to Maximize SARS-CoV-2 Case-Finding: Evaluation and Feasibility Study (Preprint)
BACKGROUND The Flexible Adaptive Algorithmic Surveillance Testing (FAAST) program represents an innovative approach for improving the detection of new cases of infectious disease; it is deployed here to screen and diagnose SARS-CoV-2. With the advent of treatment for COVID-19, finding individuals infected with SARS-CoV-2 is an urgent clinical and public health priority. While these kinds of Bayesian search algorithms are used widely in other settings (eg, to find downed aircraft, in submarine recovery, and to aid in oil exploration), this is the first time that Bayesian adaptive approaches have been used for active disease surveillance in the field. OBJECTIVE This study’s objective was to evaluate a Bayesian search algorithm to target hotspots of SARS-CoV-2 transmission in the community with the goal of detecting the most cases over time across multiple locations in Columbus, Ohio, from August to October 2021. METHODS The algorithm used to direct pop-up SARS-CoV-2 testing for this project is based on Thompson sampling, in which the aim is to maximize the average number of new cases of SARS-CoV-2 diagnosed among a set of testing locations based on sampling from prior probability distributions for each testing site. An academic-governmental partnership between Yale University, The Ohio State University, Wake Forest University, the Ohio Department of Health, the Ohio National Guard, and the Columbus Metropolitan Libraries conducted a study of bandit algorithms to maximize the detection of new cases of SARS-CoV-2 in this Ohio city in 2021. The initiative established pop-up COVID-19 testing sites at 13 Columbus locations, including library branches, recreational and community centers, movie theaters, homeless shelters, family services centers, and community event sites. Our team conducted between 0 and 56 tests at the 16 testing events, with an overall average of 25.3 tests conducted per event and a moving average that increased over time. Small incentives—including gift cards and take-home rapid antigen tests—were offered to those who approached the pop-up sites to encourage their participation. RESULTS Over time, as expected, the Bayesian search algorithm directed testing efforts to locations with higher yields of new diagnoses. Surprisingly, the use of the algorithm also maximized the identification of cases among minority residents of underserved communities, particularly African Americans, with the pool of participants overrepresenting these people relative to the demographic profile of the local zip code in which testing sites were located. CONCLUSIONS This study demonstrated that a pop-up testing strategy using a bandit algorithm can be feasibly deployed in an urban setting during a pandemic. It is the first real-world use of these kinds of algorithms for disease surveillance and represents a key step in evaluating the effectiveness of their use in maximizing the detection of undiagnosed cases of SARS-CoV-2 and other infections, such as HIV.
- Open Access
- 27 December 2018
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 13 (12) , e0209488
Veterans’ compensation claims beliefs predict timing of PTSD treatment use relative to compensation and pension exam
In this study we developed the Disability Beliefs Scale to assess Veterans’ beliefs that engaging in treatment, as well as other behaviors, would affect the likelihood of a Veteran’s being awarded disability-related benefits. We posited that Veterans with stronger beliefs that attending mental health treatment would facilitate a service-connection award would be more likely to attend PTSD treatment before their compensation and pension examinations for PTSD. Electronic health records for 307 post-9/11-era Veterans applying for compensation and pension for service-connected PTSD and engaging in a clinical trial of a treatment-referral intervention were analyzed for PTSD-specific and more general mental health treatment use around the time of their compensation examinations. All participants completed the Disability Beliefs Scale and other baseline assessments. Multilevel models assessed change in treatment use as a function of time relative to the C&P exam, compensation examination status (before or after), and the interaction between examination status and beliefs about treatment benefits. No main effects of time or examination status were observed. As hypothesized, beliefs about treatment benefits moderated the effect of examination status on PTSD treatment use. Veterans believing more strongly that mental health treatment would help a claim differentially attended PTSD treatment before the examination than after. The effect was not observed for general mental health treatment use. The association between Veterans’ use of PTSD treatment and their service-connection examination status was moderated by beliefs that receiving treatment affects the service-connection decision. This suggests that factors reported to motivate seeking service-connection—finances, validation of Veterans’ experiences, and the involvement of significant others—might also help motivate Veterans’ use of effective PTSD treatments. However, the results reflect correlations that could be explained in other ways, and service-connection was one of many factors impacting PTSD treatment engagement.
- 27 December 2018
- Published by Public Library of Science (PLoS) in PLOS ONE
- Vol. 13 (12) , e0209488
Open AccessVeterans’ compensation claims beliefs predict timing of PTSD treatment use relative to compensation and pension exam
In this study we developed the Disability Beliefs Scale to assess Veterans’ beliefs that engaging in treatment, as well as other behaviors, would affect the likelihood of a Veteran’s being awarded disability-related benefits. We posited that Veterans with stronger beliefs that attending mental health treatment would facilitate a service-connection award would be more likely to attend PTSD treatment before their compensation and pension examinations for PTSD. Electronic health records for 307 post-9/11-era Veterans applying for compensation and pension for service-connected PTSD and engaging in a clinical trial of a treatment-referral intervention were analyzed for PTSD-specific and more general mental health treatment use around the time of their compensation examinations. All participants completed the Disability Beliefs Scale and other baseline assessments. Multilevel models assessed change in treatment use as a function of time relative to the C&P exam, compensation examination status (before or after), and the interaction between examination status and beliefs about treatment benefits. No main effects of time or examination status were observed. As hypothesized, beliefs about treatment benefits moderated the effect of examination status on PTSD treatment use. Veterans believing more strongly that mental health treatment would help a claim differentially attended PTSD treatment before the examination than after. The effect was not observed for general mental health treatment use. The association between Veterans’ use of PTSD treatment and their service-connection examination status was moderated by beliefs that receiving treatment affects the service-connection decision. This suggests that factors reported to motivate seeking service-connection—finances, validation of Veterans’ experiences, and the involvement of significant others—might also help motivate Veterans’ use of effective PTSD treatments. However, the results reflect correlations that could be explained in other ways, and service-connection was one of many factors impacting PTSD treatment engagement.
- 3 April 2018
- Published by Taylor & Francis in Journal of Dual Diagnosis
- Vol. 14 (2) , 130-136
Financial Capability: Clinicians' Assessment of Beneficiaries With Dual Diagnoses
Objective: Adults with mental health disorders whose ability to work is sufficiently impeded are entitled to financial supports from the Social Security Administration. Beneficiaries determined to be incapable of managing these funds are supposed to be assigned a representative payee to assist beneficiaries in meeting their needs. However, patterns of payee assignment suggest that payee assignment is impacted by factors other than those the Social Security Administration instructs clinicians to consider. In this study, we tested the association between clinicians' judgments of their clients' financial capability and hypothesized predictors (demographic characteristics, psychiatric diagnosis, recent alcohol and other substance use, self-rated money mismanagement, recent homelessness, and provider characteristics). We posited that predictors might act indirectly on capability judgment via their impact on beneficiaries' money management. Methods: Altogether, 261 people receiving intensive mental health care who did not have payees or fiduciaries were enrolled after providing written informed consent. These beneficiaries completed in-person assessment interviews, reporting demographic characteristics, treatment history, substance use, and homelessness. Mental health clinicians identified by the beneficiaries were enrolled in the study and asked to judge their clients' financial capability with standard Social Security instructions for determining capability. Bivariate associations between hypothesized predictors and clinicians' determinations of incapability were tested. In multivariate probit regression models, incapability determination was modeled as a function of all beneficiary and clinician characteristics that had significant bivariate associations with the outcome. Results: Providers identified 24% of their clients as financially incapable. Determinations of financial incapability were unrelated to any beneficiary or provider demographic characteristics but were positively associated with money mismanagement, homelessness, and having a psychotic disorder. Alcohol use and other substance use were only significantly associated with capability determinations indirectly through their effects on money mismanagement. Conclusions: Providers' judgments of beneficiaries' capability to manage their funds were associated with factors that were consistent with Social Security Administration guidelines and were, importantly, not associated with personal characteristics. This finding suggests that guidelines can be fairly applied by clinicians and that reported inconsistencies in payee assignment are accounted for by other factors. The Social Security Administration is currently considering other approaches to standardize capability determinations.
- 3 April 2018
- Published by Taylor & Francis in Journal of Dual Diagnosis
- Vol. 14 (2) , 130-136
Financial Capability: Clinicians' Assessment of Beneficiaries With Dual Diagnoses
Objective: Adults with mental health disorders whose ability to work is sufficiently impeded are entitled to financial supports from the Social Security Administration. Beneficiaries determined to be incapable of managing these funds are supposed to be assigned a representative payee to assist beneficiaries in meeting their needs. However, patterns of payee assignment suggest that payee assignment is impacted by factors other than those the Social Security Administration instructs clinicians to consider. In this study, we tested the association between clinicians' judgments of their clients' financial capability and hypothesized predictors (demographic characteristics, psychiatric diagnosis, recent alcohol and other substance use, self-rated money mismanagement, recent homelessness, and provider characteristics). We posited that predictors might act indirectly on capability judgment via their impact on beneficiaries' money management. Methods: Altogether, 261 people receiving intensive mental health care who did not have payees or fiduciaries were enrolled after providing written informed consent. These beneficiaries completed in-person assessment interviews, reporting demographic characteristics, treatment history, substance use, and homelessness. Mental health clinicians identified by the beneficiaries were enrolled in the study and asked to judge their clients' financial capability with standard Social Security instructions for determining capability. Bivariate associations between hypothesized predictors and clinicians' determinations of incapability were tested. In multivariate probit regression models, incapability determination was modeled as a function of all beneficiary and clinician characteristics that had significant bivariate associations with the outcome. Results: Providers identified 24% of their clients as financially incapable. Determinations of financial incapability were unrelated to any beneficiary or provider demographic characteristics but were positively associated with money mismanagement, homelessness, and having a psychotic disorder. Alcohol use and other substance use were only significantly associated with capability determinations indirectly through their effects on money mismanagement. Conclusions: Providers' judgments of beneficiaries' capability to manage their funds were associated with factors that were consistent with Social Security Administration guidelines and were, importantly, not associated with personal characteristics. This finding suggests that guidelines can be fairly applied by clinicians and that reported inconsistencies in payee assignment are accounted for by other factors. The Social Security Administration is currently considering other approaches to standardize capability determinations.
- 1 November 2013
- Published by Elsevier in NeuroImage
- Vol. 81, 110-118
Real-time fMRI links subjective experience with brain activity during focused attention
- 1 November 2013
- Published by Elsevier in NeuroImage
- Vol. 81, 110-118
Real-time fMRI links subjective experience with brain activity during focused attention
- 1 June 2013
- Published by Elsevier in Drug and Alcohol Dependence
- Vol. 130 (1-3) , 222-229
Mindfulness training for smoking cessation: Moderation of the relationship between craving and cigarette use
- 1 June 2013
- Published by Elsevier in Drug and Alcohol Dependence
- Vol. 130 (1-3) , 222-229
Mindfulness training for smoking cessation: Moderation of the relationship between craving and cigarette use
- Open Access
- 1 January 2013
- Published by Frontiers Media SA in Frontiers in Human Neuroscience
- Vol. 7, 440
Effortless awareness: using real time neurofeedback to investigate correlates of posterior cingulate cortex activity in meditators' self-report
Neurophenomenological studies seek to utilize first-person self-report to elucidate cognitive processes related to physiological data. Grounded theory offers an approach to the qualitative analysis of self-report, whereby theoretical constructs are derived from empirical data. Here we used grounded theory methodology (GTM) to assess how the first-person experience of meditation relates to neural activity in a core region of the default mode network—the posterior cingulate cortex (PCC). We analyzed first-person data consisting of meditators' accounts of their subjective experience during runs of a real time fMRI neurofeedback study of meditation, and third-person data consisting of corresponding feedback graphs of PCC activity during the same runs. We found that for meditators, the subjective experiences of “undistracted awareness” such as “concentration” and “observing sensory experience,” and “effortless doing” such as “observing sensory experience,” “not efforting,” and “contentment,” correspond with PCC deactivation. Further, the subjective experiences of “distracted awareness” such as “distraction” and “interpreting,” and “controlling” such as “efforting” and “discontentment,” correspond with PCC activation. Moreover, we derived several novel hypotheses about how specific qualities of cognitive processes during meditation relate to PCC activity, such as the difference between meditation and “trying to meditate.” These findings offer novel insights into the relationship between meditation and mind wandering or self-related thinking and neural activity in the default mode network, driven by first-person reports.
- 1 January 2013
- Published by Frontiers Media SA in Frontiers in Human Neuroscience
- Vol. 7, 440
Open AccessEffortless awareness: using real time neurofeedback to investigate correlates of posterior cingulate cortex activity in meditators' self-report
Neurophenomenological studies seek to utilize first-person self-report to elucidate cognitive processes related to physiological data. Grounded theory offers an approach to the qualitative analysis of self-report, whereby theoretical constructs are derived from empirical data. Here we used grounded theory methodology (GTM) to assess how the first-person experience of meditation relates to neural activity in a core region of the default mode network—the posterior cingulate cortex (PCC). We analyzed first-person data consisting of meditators' accounts of their subjective experience during runs of a real time fMRI neurofeedback study of meditation, and third-person data consisting of corresponding feedback graphs of PCC activity during the same runs. We found that for meditators, the subjective experiences of “undistracted awareness” such as “concentration” and “observing sensory experience,” and “effortless doing” such as “observing sensory experience,” “not efforting,” and “contentment,” correspond with PCC deactivation. Further, the subjective experiences of “distracted awareness” such as “distraction” and “interpreting,” and “controlling” such as “efforting” and “discontentment,” correspond with PCC activation. Moreover, we derived several novel hypotheses about how specific qualities of cognitive processes during meditation relate to PCC activity, such as the difference between meditation and “trying to meditate.” These findings offer novel insights into the relationship between meditation and mind wandering or self-related thinking and neural activity in the default mode network, driven by first-person reports.