1
|
Du S, Yao J, Shen GC, Lin B, Udo T, Hastings J, Wang F, Wang F, Zhang Z, Ye X, Zhang K. Social Drivers of Mental Health: A U.S. Study Using Machine Learning. Am J Prev Med 2023; 65:827-834. [PMID: 37286016 DOI: 10.1016/j.amepre.2023.05.022] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 02/23/2023] [Revised: 05/26/2023] [Accepted: 05/30/2023] [Indexed: 06/09/2023]
Abstract
INTRODUCTION Social drivers of mental health can be compared on an aggregated level. This study employed a machine learning approach to identify and rank social drivers of mental health across census tracts in the U.S. METHODS Data for 38,379 census tracts in the U.S. were collected from multiple sources in 2021. Two measures of mental health problems-self-reported depression and self-assessed poor mental health-among adults and three domains of social drivers (behavioral, environmental, and social) were analyzed on the basis of the unit of census tracts using the Extreme Gradient Boosting machine learning approach in 2022. The leading social drivers were found in each domain in the main sample and in the subsamples divided on the basis of poverty and racial segregation. RESULTS The three domains combined explained more than 90% of the variance of both mental illness indicators. Self-reported depression and self-assessed poor mental health differed in major social drivers. The two outcome indicators had one overlapping correlate from the behavioral domain: smoking. Other than smoking, climate zone and racial composition were the leading correlates from the environmental and social domains, respectively. Census tract characteristics moderated the impacts of social drivers on mental health problems; the major social drivers differed by census tract poverty and racial segregation. CONCLUSIONS Population mental health is highly contextualized. Better interventions can be developed on the basis of census tract-level analyses of social drivers that characterize the upstream causes of mental health problems.
Collapse
Affiliation(s)
- Shichao Du
- Department of Sociology, University at Albany, State University of New York, Albany, New York
| | - Jie Yao
- Department of Epidemiology & Biostatistics, School of Public Health, University at Albany, State University of New York, Rensselaer, New York
| | - Gordon C Shen
- Department of Management, Policy & Community Health, School of Public Health, The University of Texas Health Science Center at Houston, Houston, Texas
| | - Betty Lin
- Department of Psychology, College of Arts and Sciences, University at Albany, State University of New York, Albany, New York
| | - Tomoko Udo
- Department of Epidemiology & Biostatistics, School of Public Health, University at Albany, State University of New York, Rensselaer, New York; Department of Health Policy, Management & Behavior, School of Public Health, University at Albany, State University of New York, Rensselaer, New York
| | - Julia Hastings
- Department of Health Policy, Management & Behavior, School of Public Health, University at Albany, State University of New York, Rensselaer, New York
| | - Fei Wang
- Institute of Artificial Intelligence for Digital Health, Department of Population Health Sciences, Weill Cornell Medicine, Cornell University, New York, New York; Division of Health Informatics, Department of Population Health Sciences, Weill Cornell Medicine, Cornell University, New York, New York
| | - Fusheng Wang
- Department of Biomedical Informatics, School of Medicine and College of Engineering and Applied Sciences, Stony Brook University, Stony Brook, New York; Department of Computer Science, Stony Brook University, Stony Brook, New York
| | - Zhe Zhang
- Department of Geography, Texas A&M University, College Station, Texas
| | - Xinyue Ye
- Department of Landscape Architecture & Urban Planning, Texas A&M University, College Station, Texas
| | - Kai Zhang
- Department of Environmental Health Sciences, School of Public Health, University of Albany, State University of New York, Rensselaer, New York.
| |
Collapse
|
2
|
Sun F, Yao J, Du S, Qian F, Appleton AA, Tao C, Xu H, Liu L, Dai Q, Joyce BT, Nannini DR, Hou L, Zhang K. Social Determinants, Cardiovascular Disease, and Health Care Cost: A Nationwide Study in the United States Using Machine Learning. J Am Heart Assoc 2023; 12:e027919. [PMID: 36802713 PMCID: PMC10111459 DOI: 10.1161/jaha.122.027919] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 02/23/2023]
Abstract
Background Existing studies on cardiovascular diseases (CVDs) often focus on individual-level behavioral risk factors, but research examining social determinants is limited. This study applies a novel machine learning approach to identify the key predictors of county-level care costs and prevalence of CVDs (including atrial fibrillation, acute myocardial infarction, congestive heart failure, and ischemic heart disease). Methods and Results We applied the extreme gradient boosting machine learning approach to a total of 3137 counties. Data are from the Interactive Atlas of Heart Disease and Stroke and a variety of national data sets. We found that although demographic composition (eg, percentages of Black people and older adults) and risk factors (eg, smoking and physical inactivity) are among the most important predictors for inpatient care costs and CVD prevalence, contextual factors such as social vulnerability and racial and ethnic segregation are particularly important for the total and outpatient care costs. Poverty and income inequality are the major contributors to the total care costs for counties that are in nonmetro areas or have high segregation or social vulnerability levels. Racial and ethnic segregation is particularly important in shaping the total care costs for counties with low poverty rates or social vulnerability level. Demographic composition, education, and social vulnerability are consistently important across different scenarios. Conclusions The findings highlight the differences in predictors for different types of CVD cost outcomes and the importance of social determinants. Interventions directed toward areas that have been economically and socially marginalized may aid in reducing the impact of CVDs.
Collapse
Affiliation(s)
- Feinuo Sun
- Global Aging and Community Initiative Mount Saint Vincent University Halifax Nova Scotia Canada
| | - Jie Yao
- Department of Epidemiology and Biostatistics, School of Public Health University at Albany, State University of New York Albany NY
| | - Shichao Du
- Department of Sociology University at Albany, State University of New York Albany NY
| | - Feng Qian
- Department of Health Policy, Management and Behavior, School of Public Health University at Albany, State University of New York Albany NY
| | - Allison A Appleton
- Department of Epidemiology and Biostatistics, School of Public Health University at Albany, State University of New York Albany NY
| | - Cui Tao
- School of Biomedical Informatics The University of Texas Health Science Center at Houston Houston TX
| | - Hua Xu
- School of Biomedical Informatics The University of Texas Health Science Center at Houston Houston TX
| | - Lei Liu
- Division of Biostatistics Washington University in St. Louis St. Louis MO
| | - Qi Dai
- Division of Epidemiology, Department of Medicine, Vanderbilt Epidemiology Center, School of Medicine Vanderbilt University, Vanderbilt-Ingram Cancer Center, Vanderbilt University Medical Center Nashville TN
| | - Brian T Joyce
- Department of Preventive Medicine Northwestern University Feinberg School of Medicine Chicago IL
| | - Drew R Nannini
- Department of Preventive Medicine Northwestern University Feinberg School of Medicine Chicago IL
| | - Lifang Hou
- Department of Preventive Medicine Northwestern University Feinberg School of Medicine Chicago IL
| | - Kai Zhang
- Department of Environmental Health Sciences, School of Public Health University at Albany, State University of New York Albany NY
| |
Collapse
|
3
|
Makridis CA, Strebel T, Marconi V, Alterovitz G. Designing COVID-19 mortality predictions to advance clinical outcomes: Evidence from the Department of Veterans Affairs. BMJ Health Care Inform 2021; 28:bmjhci-2020-100312. [PMID: 34108143 PMCID: PMC8190987 DOI: 10.1136/bmjhci-2020-100312] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 12/23/2020] [Revised: 03/17/2021] [Accepted: 03/31/2021] [Indexed: 12/21/2022] Open
Abstract
Using administrative data on all Veterans who enter Department of Veterans Affairs (VA) medical centres throughout the USA, this paper uses artificial intelligence (AI) to predict mortality rates for patients with COVID-19 between March and August 2020. First, using comprehensive data on over 10 000 Veterans’ medical history, demographics and lab results, we estimate five AI models. Our XGBoost model performs the best, producing an area under the receive operator characteristics curve (AUROC) and area under the precision-recall curve of 0.87 and 0.41, respectively. We show how focusing on the performance of the AUROC alone can lead to unreliable models. Second, through a unique collaboration with the Washington D.C. VA medical centre, we develop a dashboard that incorporates these risk factors and the contributing sources of risk, which we deploy across local VA medical centres throughout the country. Our results provide a concrete example of how AI recommendations can be made explainable and practical for clinicians and their interactions with patients.
Collapse
Affiliation(s)
- Christos A Makridis
- National Artificial Intelligence Institute at the Department of Veterans Affairs, US Department of Veterans Affairs, Washington, District of Columbia, USA .,Digital Economy Lab, Stanford University, Stanford University, Stanford, California, USA
| | - Tim Strebel
- Washington D.C. VA Medical Center, Department of Veterans Affairs, Washington, District of Columbia, USA
| | - Vincent Marconi
- Rollins School of Public Health, Emory University, Atlanta, Georgia, USA
| | - Gil Alterovitz
- National Artificial Intelligence Institute at the Department of Veterans Affairs, US Department of Veterans Affairs, Washington, District of Columbia, USA.,Harvard Medical School, Boston, Massachusetts, USA
| |
Collapse
|
4
|
Makridis C, Hurley S, Klote M, Alterovitz G. Ethical Applications of Artificial Intelligence: Evidence From Health Research on Veterans. JMIR Med Inform 2021; 9:e28921. [PMID: 34076584 PMCID: PMC8209529 DOI: 10.2196/28921] [Citation(s) in RCA: 4] [Impact Index Per Article: 1.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/18/2021] [Revised: 04/23/2021] [Accepted: 04/27/2021] [Indexed: 01/01/2023] Open
Abstract
Background Despite widespread agreement that artificial intelligence (AI) offers significant benefits for individuals and society at large, there are also serious challenges to overcome with respect to its governance. Recent policymaking has focused on establishing principles for the trustworthy use of AI. Adhering to these principles is especially important for ensuring that the development and application of AI raises economic and social welfare, including among vulnerable groups and veterans. Objective We explore the newly developed principles around trustworthy AI and how they can be readily applied at scale to vulnerable groups that are potentially less likely to benefit from technological advances. Methods Using the US Department of Veterans Affairs as a case study, we explore the principles of trustworthy AI that are of particular interest for vulnerable groups and veterans. Results We focus on three principles: (1) designing, developing, acquiring, and using AI so that the benefits of its use significantly outweigh the risks and the risks are assessed and managed; (2) ensuring that the application of AI occurs in well-defined domains and is accurate, effective, and fit for the intended purposes; and (3) ensuring that the operations and outcomes of AI applications are sufficiently interpretable and understandable by all subject matter experts, users, and others. Conclusions These principles and applications apply more generally to vulnerable groups, and adherence to them can allow the VA and other organizations to continue modernizing their technology governance, leveraging the gains of AI while simultaneously managing its risks.
Collapse
Affiliation(s)
- Christos Makridis
- National Artificial Intelligence Institute, Department of Veterans Affairs, Washington, DC, United States.,Stanford Digital Economy Lab, Stanford University, Stanford, CA, United States.,WP Carey School of Business, Arizona State University, Tempe, AZ, United States
| | - Seth Hurley
- National Artificial Intelligence Institute, Department of Veterans Affairs, Washington, DC, United States
| | - Mary Klote
- Office of Research & Development, Department of Veterans Affairs, Washington, DC, United States
| | - Gil Alterovitz
- National Artificial Intelligence Institute, Department of Veterans Affairs, Washington, DC, United States.,Boston Children's Hospital, Harvard Medical School, Boston, MA, United States
| |
Collapse
|