Off-campus UMass Amherst users: To download campus access dissertations, please use the following link to log into our proxy server with your UMass Amherst user name and password.

Non-UMass Amherst users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

Dissertations that have an embargo placed on them will not be available to anyone until the embargo expires.


Access Type

Open Access Thesis

Document Type


Degree Program

Public Health

Degree Type

Master of Science (M.S.)

Year Degree Awarded


Month Degree Awarded



Higher levels of allostatic load (AL), a composite measure of the impact of chronic stress on the body, are found among socially marginalized groups compared to privileged groups. AL is associated with premature aging and death, as well as a variety of chronic health conditions that impact quality of life. Effects of AL may be offset by physical activity (PA). Queer populations (including those who identify as lesbian, gay, or bisexual) may be at risk for elevated AL due to repeated exposure to discrimination in the form of sexual minority stress. There is mixed literature on sexual orientation differences in PA and research on AL among queer populations is limited. This study describes sexual orientation differences in the 1) prevalence of PA and mean AL levels and 2) association between PA and AL. Data from the 2001-2015 National Health and Nutrition Examination Survey were used to fit sex-stratified linear regression models assessing sexual orientation (heterosexual, gay/lesbian, bisexual, and heterosexual with same-sex experience) differences as well as sexual orientation-specific associations in PA and AL. AL was assessed using 8 biomarkers, while PA was assessed using metabolic equivalent of task (MET)-hours/week calculated from the Global Physical Activity Questionnaire. Models were adjusted for race/ethnicity, income, education, relationship status, veteran status, and citizenship status. Overall, heterosexual adults reported more PA than their same-sex queer counterparts. Gay men had lower (0.8), while bisexual men had higher AL scores (1.9) than heterosexual men (1.2). Among women, AL score was similar across sexual orientation groups. Overall, more PA was associated with lower AL scores among men and women (badjusted = -0.00508, padjusted=-0.00505, punadjusted =-0.00934, punadjusted =-0.03269, p=0.04), and heterosexual women with same-sex experience (bunadjusted =-0.01059, p=0.02). After adjusting for covariates the association was no longer statistically significant. Consistent with previous work, PA was inversely associated with AL score. PA may be an important modifiable behavior to mitigate some sexual orientation disparities in health.


First Advisor

Nicole VanKim

Second Advisor

Susan Hankinson

Included in

Epidemiology Commons