xiange wang, Wenhuan Tan, Kaitlyn Martinez, Benjamin H. McMahon, Jean C. Beckham, Nathan A. Kimbrel, Silvia Crivelli
{"title":"Association between social and environmental determinants of health with suicide-related death among veterans","authors":"xiange wang, Wenhuan Tan, Kaitlyn Martinez, Benjamin H. McMahon, Jean C. Beckham, Nathan A. Kimbrel, Silvia Crivelli","doi":"10.1101/2024.07.02.24309854","DOIUrl":null,"url":null,"abstract":"<strong>Importance</strong> Social and environmental determinants of health (SDOH and EDOH) may contribute significantly to suicide rates among U.S. veterans. <strong>Objective</strong> To identify key predictive variables for assessing suicide related death rates (SRR), which include suicide deaths, suicide firearm deaths, and suicide nonfirearm deaths and vulnerability areas. <strong>Design, Setting, and Participants</strong> This case control study utilized Electronic Health Record (EHR) data, which included demographic and mental health information spanning from January 1, 2006, to December 31, 2016. The base cohort considered all veterans from the VHA outpatient database during the above period. Patients from the base cohort who died by suicide were identified through the National Death Index and considered as cases. Given the significantly larger number of alive patients compared to deceased patients, which caused the dataset to be extremely unbalanced and potentially biased, control participants were selected at a ratio of 4 controls to 1 case from those who were still alive. Cases of suicide related death were matched with four controls based on birth year, cohort entry date, sex, and follow up duration. Comprehensive data on social determinants (SDOH), geographic and gun related factors, quality of access to healthcare, environmental determinants (EDOH), and food insecurity were gathered from various sources at the midpoint of the study in 2011. Data analysis was carried out from January 2023 to January 2024.\n<strong>Exposures</strong> Suicide related deaths associated with SDOH and EDOH.\n<strong>Main Outcomes and Measures</strong> A hierarchical clustering method was employed to downselect the large number of variables, while Cox regression models were used to identify key predictive variables for SRR and areas of vulnerability.\n<strong>Results</strong> Out of a total of 9,819,080 veterans, 28,302 were identified as having died by suicide. These cases were matched with 113,208 control participants. The majority of the cohort was male (137,264 [97%]) and White (101,533 [72%]), with a significant portion being Black veterans (18,450 [13.12%]). The average age (SD) was 64.77 (17.56) years. We found that Social Determinants of Health (SDOH) and Environmental Determinants of Health (EDOH) were significantly associated with an increased risk of suicide. By incorporating SDOH and EDOH into the model, the performance (AUC) improved from 0.70 to 0.73. <strong>Conclusions and Relevance</strong> In this study, veterans who died by suicide using firearms exhibited distinct characteristics based on SDOH and EDOH, particularly in gun related variables, compared to those who died by nonfirearm methods. Our analysis indicated that veterans living in areas with more social issues, higher temperatures, and higher altitudes are at a higher risk of all means suicide. Furthermore, regions such as Montana, Wyoming, West Virgina and Arkansas, characterized by higher gun owernship are predicted to have the highest vulnerability based on veteran suicide firearm rates. Gun ownership and gun laws grades showed as strong predictors rather than rurality.","PeriodicalId":501072,"journal":{"name":"medRxiv - Health Economics","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-07-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"medRxiv - Health Economics","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1101/2024.07.02.24309854","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Importance Social and environmental determinants of health (SDOH and EDOH) may contribute significantly to suicide rates among U.S. veterans. Objective To identify key predictive variables for assessing suicide related death rates (SRR), which include suicide deaths, suicide firearm deaths, and suicide nonfirearm deaths and vulnerability areas. Design, Setting, and Participants This case control study utilized Electronic Health Record (EHR) data, which included demographic and mental health information spanning from January 1, 2006, to December 31, 2016. The base cohort considered all veterans from the VHA outpatient database during the above period. Patients from the base cohort who died by suicide were identified through the National Death Index and considered as cases. Given the significantly larger number of alive patients compared to deceased patients, which caused the dataset to be extremely unbalanced and potentially biased, control participants were selected at a ratio of 4 controls to 1 case from those who were still alive. Cases of suicide related death were matched with four controls based on birth year, cohort entry date, sex, and follow up duration. Comprehensive data on social determinants (SDOH), geographic and gun related factors, quality of access to healthcare, environmental determinants (EDOH), and food insecurity were gathered from various sources at the midpoint of the study in 2011. Data analysis was carried out from January 2023 to January 2024.
Exposures Suicide related deaths associated with SDOH and EDOH.
Main Outcomes and Measures A hierarchical clustering method was employed to downselect the large number of variables, while Cox regression models were used to identify key predictive variables for SRR and areas of vulnerability.
Results Out of a total of 9,819,080 veterans, 28,302 were identified as having died by suicide. These cases were matched with 113,208 control participants. The majority of the cohort was male (137,264 [97%]) and White (101,533 [72%]), with a significant portion being Black veterans (18,450 [13.12%]). The average age (SD) was 64.77 (17.56) years. We found that Social Determinants of Health (SDOH) and Environmental Determinants of Health (EDOH) were significantly associated with an increased risk of suicide. By incorporating SDOH and EDOH into the model, the performance (AUC) improved from 0.70 to 0.73. Conclusions and Relevance In this study, veterans who died by suicide using firearms exhibited distinct characteristics based on SDOH and EDOH, particularly in gun related variables, compared to those who died by nonfirearm methods. Our analysis indicated that veterans living in areas with more social issues, higher temperatures, and higher altitudes are at a higher risk of all means suicide. Furthermore, regions such as Montana, Wyoming, West Virgina and Arkansas, characterized by higher gun owernship are predicted to have the highest vulnerability based on veteran suicide firearm rates. Gun ownership and gun laws grades showed as strong predictors rather than rurality.