PAUL ET AL. RESPOND

We are gratified to note that Mathy does not have any substantive criticisms of our research. However, we would like to address 2 of Mathy’s points: first, that other population-based studies on this topic have been conducted, and second, that our analyses lacked cultural sensitivity. As our comment that prior research has been “primarily based on opportunistic samples” does not deny that some population-based research has occurred (which we cite), I am at pains to understand the contention that our “assertion” is incorrect. The implication is that there are no benefits to the publication of our article. This can be readily rebutted. Our analyses were based on data from a rigorously derived population-based sample of men who have sex with men (MSM) in 4 major US cities. The representative nature and size of this sample, along with the breadth of health issues covered in telephone interviews, have offered a unique perspective on urban MSM. Such a perspective is needed in addition to the groundbreaking work with opportunistic samples that typically preceded this research. Mathy cited Remafedi, who identified 6 “population-based and controlled studies published since 1997.”1 These publications actually represent 5 studies, as 2 articles2,3 reported on the same data set. Four2–6 were studies of school-based samples of adolescents, using the Youth Risk Behavior Survey. Variations in the operationalization of sexual orientation (e.g., self-label, sexual behavior, “romantic attractions”) not only meant variations in the comparison groups but have led to discrepant findings in other analyses.7,8 Further, we noted that adolescents who self-identify as gay, lesbian, or bisexual (GLB) by 13 to 18 years of age may not be representative of the broader GLB population, a point echoed by others. The remaining study defined a population-based sample of 750 men aged 18 to 27 years in a neighborhood in Calgary, Canada.9 However, the reported prevalence of homosexual activity, identity, or both contradicts prior demographic research and suggests the need for confirmatory research.10 As we noted, subsequent studies’ analyses using national data sets have been limited by relatively small numbers of MSM and lack of gay-specific variables. Mathy claims that our age cohort analysis lacked sensitivity to the culturally significant health-related events of this population. While the removal of homosexuality as a disorder from the Diagnostic and Statistical Manual of Mental Disorders (DSM) in 1973 signaled a significant, affirmative shift in how the professional and academic world viewed homosexuality, it was embedded within larger social changes. It is unclear how immediately this action influenced the developmental experiences of gay men, lesbians, and bisexuals. The professional and academic resistance to this change was evidenced by the necessity 14 years later of removing “ego-dystonic homosexuality” from the DSM. Mathy’s critique does an injustice to the degree to which cultural context was considered in our article. We used a key watershed event—the Stonewall Riots and the birth of the modern gay liberation movement—as this marked an era of social transformation and empowerment for the GLB community. Research on GLB youth in the United States since that date (e.g., Cohen and Savin-Williams11) suggests that marked changes have occurred over time, not necessarily linked to any one event but as part of a progressive cultural shift. The reactions to these social changes are complex; evidence also suggests that GLB youth are facing greater violence today than in decades past.12 Thus we chose to compare those who reached age 25 by 1970 with cohorts that reached that age within subsequent decades, a 10-year span being not uncommon to this type of analysis. Our cohort analyses considered one marker in the growth of the GLB community, rather than other milestones suggested by Mathy. We may differ in the significance we attach to these milestones, but that is all.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.