![IconGems-Print](images/IconGems-Print.png)
197 – Health Data Analytics
Methods for Calculating State and National Prevalence Estimates: An Application of Estimates of Sexual Orientation and Gender Identity
Yangyang Deng
ICF
This research focuses on the methods for modeling estimates at the state level when data are available from a subset of states. We used the Sexual Orientation and Gender Identity (SOGI) optional module questions from the Behavioral Risk Factor Surveillance System (BRFSS) for 2016 to 2018 to develop models and provide estimates for all states. Models are validated against direct estimates where available. SOGI questions represent the most vigorous test of such a model in that a limited proportion of the sample who identify as transgender, bisexual and/or gay/lesbian. The process presented also provides a mechanism for imputation of responses where non-substantive answers are given (i.e. “do not know� or refusal to answer). The methodology is adaptable to other BRFSS optional models used by subsets of the states annually.