Online Program

Health outcomes of non-heterosexual compared to heterosexual adolescents and young adults from a national survey in viet nam: Analysis using propensity score matching

Monday, November 4, 2013

Trang Q. Nguyen, MS, Department of Health, Behavior and Society, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD
Elizabeth Stuart, PhD, Department of Mental Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD
Methodological background: A challenge in LGBT health disparities research is lack of comparable LGBT and non-LGBT samples. Even with population samples (preferred over separately recruited LGBT and non-LGBT samples), confounding remains problematic due to small LGBT percentages. Geographical background: No prior research examined LGBT health disparities in Viet Nam. Methods: We analyzed data from a national survey of adolescents and young adults in Viet Nam. 160 male and 155 female participants reported being attracted to the same or both genders. These non-heterosexual individuals were matched to heterosexual (attracted to the opposite gender) individuals on age, minority/majority ethnicity, urban/rural location, regions, education, work status, family assets, and parents' education. Variable ratio matching was used. Matching combined propensity score with caliper, Mahalanobis distance on continuous variables, and exact matching on important categorical variables, using the R-package MatchIt. We matched in several steps to acquire matches for all non-heterosexual individuals. Risk ratios of various health outcomes were evaluated using matched samples. Results: 1482 male and 622 female heterosexual persons were selected as matches, with excellent balance of propensity score and all characteristics. Non-heterosexual females are more likely than heterosexual females to smoke (RR=20, 95% CI 1.25-318) and to have been injured by family members (RR=4.08, CI 1.79-9.30). Non-heterosexual males are more likely than heterosexual males to have thought about suicide (RR=3.78, CI 1.75-8.13). Both non-heterosexual males and females are more likely than heterosexuals to have been very depressed (RR=1.40, CI 1.08-1.82, and RR=1.36, CI 1.11-1.65, respectively).

Learning Areas:

Biostatistics, economics
Public health or related research
Social and behavioral sciences

Learning Objectives:
Describe how to analyze population data to identify health disparities borne by LGBT by using matching techniques to achieve samples of LGBT and non-LGBT persons that are comparable on a defined list of characteristics, as a way to remove confounding

Keyword(s): Health Disparities, Sexuality

Presenting author's disclosure statement:

Qualified on the content I am responsible for because: I am a doctoral candidate in social and behavior sciences and health. Among my research interests are LGBT health and rights, HIV/AIDS, social determinants of health, and statistical methods.
Any relevant financial relationships? No

I agree to comply with the American Public Health Association Conflict of Interest and Commercial Support Guidelines, and to disclose to the participants any off-label or experimental uses of a commercial product or service discussed in my presentation.

Back to: 3303.0: LGBT health poster session