|
Abstract
|
Reviews of empirical studies examining the health of gay, lesbian and bisexual individuals (GLB) led to contradictory findings about the prevalence of health problems in this population. These studies employed non probability sampling and suffered from a number of methodological shortcomings that could explain the contradictory findings. Given that many recently published studies on GLB health employed probability sampling and used health measures with good psychometric properties, a comprehensive review of these studies could help to draw a more coherent picture of GLB individuals' health.
OBJECTIVEs. a) To catalogue studies that employed probability sampling and compared the health of individuals of different sexual orientations; b) to describe methodological characteristics of these studies; and c) to examine the convergence of the findings regarding GLB health.
METHOD. Using medical sciences, psychology and sociology databases, 32 publications were identified and catalogued using a coding system enabling content, method, and metric properties classification. Two independent judges conducted the analyses of the documents.
RESULTS. As compared to studies not using probability sampling, studies using probability sampling had fewer methodological shortcomings and more coherent findings on GLB health. There were higher proportions of GLB individuals, relative to heterosexual individuals, who reported suicide attempts, high levels of psychological distress, substance abuse, and incidence of victimization.
CONCLUSION. The probability sample studies on GLB health showed robust findings suggesting that the GLB population is vulnerable. Because no empirical findings have documented that homosexuality per se is a cause of illness, future studies should identify the environmental factors likely to account for the higher prevalence of healh problems in the GLB population relative to the heterosexual population.
Language: fr |