Estimating the Population Prevalence and Risk of Suicide for the LGBTQ+ Population in the United States

Suicide research has consistently documented that lesbian, gay, bisexual, and transgender (LGBT+) people have worse mental health outcomes, including an elevated risk of suicidal ideation and attempts than their heterosexual counterparts. However, scientific progress on identifying the nature and magnitude of the risk of completed suicide for LGBT+ individuals is hindered by the lack of information on the sexual orientation and gender identity of suicide decedents as well as living individuals in population-based data. This project breaks through this gap by constructing a novel multilevel dataset for examining US completed suicides in three steps. First, I integrate individual-level data on suicide cases from the National Violent Death Reporting System (NVDRS) with data on living individuals from the Behavioral Risk Factor Surveillance System (BRFSS) data on using the geographic matching strategy that we pioneered (Pescosolido, Lee, and Kafadar 2020). This harmonization allows an examination of suicide risk associated with sexual orientation for those who are 18 years old and older. Second, known risk and protective factors for suicidal attempts in the LGBT+ population will be harmonized across two data sets. Third, concerns about missing data will be dealt with by limiting my analysis to six states (Colorado, Connecticut, Delaware, Hawaii, and Massachusetts) where missing LGBT+ status is the lowest and missing imputation models can leverage individual socio-demographic characteristics as well as other contextual variables such as place of living, relationship problems, and state-level death investigator systems. Using the harmonized data set, this project will accomplish the following two aims: 1) To determine suicide rates among the LGBT+ population in the US, and 2) To investigate how individual risk and protective factors (e.g., age, mental/physical health problems and marriage) for completed suicide in the LGBT+ population differ from their heterosexual counterparts. Accomplishing these goals will provide the first population-based estimates of LGBT+ suicide and risk/protective factors. That is, while studies examining suicide attempts and ideation show an increased risk among LGBT+ people, existing studies examining suicide deaths report contradictory findings on the elevated risk of suicide death in the LGBT+ population, perhaps because of their small cohort samples. By solving the problems of missingness in the NVDRS and the absence of data on living individuals simultaneously, this project will generate reliable estimates for the state-level prevalence of LGBT+ complete suicide. Given the known gap between factors affecting suicide attempts and completed suicides in the non-LGBT+ population, it is essential to examine whether the same gap exists in the LGBT+ population. The successful execution of the proposed work will produce a deeper understanding of suicide risk associated with sexual orientation and gender identity, and research methods to address the missing problem of LGBT status in the NVDRS data. Understanding the risk factors for suicide among the LGBT+ population net of the baseline socio-demographic difference will provide the necessary scientific foundation to design effective suicide programs and policies.