The current study used a novel measure of the social environment that did not rely on self-report perceptions and linked this measure to suicide attempts in a population-based sample of youth. Results indicated that living in environments that are less supportive of gays and lesbians is associated with greater suicide attempts among LGB youth. Previous studies have documented several factors that increase the risk for suicide attempts among LGB youth, including depression,7
hazardous alcohol use,6
and physical abuse by an adult.21
Even after adjusting for these individual-level risk factors, the social environment was associated with suicide attempts in this sample.
There also was a reduction in the association between sexual orientation and suicide attempts when the social environment and individual-level risk factors were controlled. This attenuation (42% for lesbian and gay youth and 57% for bisexual youth) was larger than that found in previous studies, which have ranged from 8%7
Although this reduction was largely driven by the individual-level risk factors (the social environment accounted for an additional 2% reduction in the association between sexual orientation and suicide attempts), on a population level these effects can have a significant public health impact. For instance, a 5-unit increase in the social-environment measure, which is plausible given the 17-point range of this measure, would lead to a 10% reduction in suicide attempts.
It is important to note, however, that LGB status remained a significant predictor of suicide attempts even after adjusting for individual-level and social-level risk factors. One possibility for the lack of full mediation is that the social climate variable was not an exhaustive index of the ecological environment for Oregon youth. Indeed, there are other contextual effects (eg, antigay attitudes) that were not included in the measure that may be associated with suicide attempts. Moreover, the OHT survey does not include measures of risk factors that are unique to LGB individuals, such as gay-related stressors34
and earlier age at disclosure,35
which are associated with suicidality. Studies that incorporate additional measures of the social context and gay-specific risk factors are needed to further test mediational hypotheses.
Future research also is needed to identify mechanisms linking aspects of harmful social environments to suicide attempts among LGB youth. One potential pathway is through increased exposure to status-based stressors, a well-documented risk factor for poor mental health in LGB populations.11
For example, LGB adults living in states in which they are denied legal protection report multiple stressors, including negative media portrayals, antigay graffiti, comments, and jokes, and a lost sense of safety.36
Stress contributes to the development of psychopathology,37,38
which in turn increases the risk of suicide attempts. Recent research also has suggested that the social environment may contribute to adverse mental health outcomes among LGB individuals through creating elevations in basic psychological risk factors for psychopathology, such as emotion regulation difficulties,39
which are associated with suicidality.40
This study has several limitations. First, the data are cross-sectional. Consequently, we cannot infer causal relationships between the social environment and suicide attempts. Prospective studies that examine how the social environment influences suicide attempts are needed to establish clearer causal inferences. Second, the OHT study only assesses youth attending schools; consequently, runaway or homeless youth are not included. Because LGB individuals are overrepresented among homeless youth,41
the OHT study may be missing a vulnerable subpopulation of LGB youth. Nevertheless, a negative social environment is likely to be a particularly robust determinant of suicide attempts among LGB homeless youth, which would have biased our results toward the null. Third, although the OHT study surveyed over one-third of youth attending public schools in Oregon, students attending private and alternative schools were not included. In addition, one-quarter of school districts declined to participate in the study. Both of these factors likely restrict the generalizability of the study's results. Future studies are therefore needed to replicate these findings using other samples of youths from diverse social contexts. Fourth, although the OHT measures have been well validated,31,32
some weaknesses of the measures have been noted, including single-item measures that may inflate the actual prevalence of suicide attempts.42
These results require replication with more detailed assessments of suicide attempts.
There also were limitations to the school policy variables. The OHT study does not release data on the specific schools that participated in the survey. The variables on school policies were aggregated across the district level and are therefore less sensitive indicators than measures of individual school policies. This likely reduced our ability to detect stronger relationships between the social environment and suicide attempts. Moreover, although the school policies represent a marker of school climate, we were unable to obtain data regarding the extent to which these policies were enforced in the schools. In addition, 18 school districts did not provide information on school policies; consequently, the school climate variables for these counties were less reliable. However, when the 4 counties with missing data were removed from the analyses, the results remained unchanged. Finally, although counties are much smaller spatial units than states, they may not reflect all aspects of community climate.43
Creating measures of ecological environments that are more proximal to LGB youth (eg, neighborhoods) will provide an opportunity to test the sensitivity of this study's results across different spatial scales,44
an important avenue for future inquiry. Nevertheless, the fact that we were able to document an association between social climate at the county-level and suicide attempts suggests that the results should be considered conservative estimates.
The current study has several noteworthy advantages for studying relationships between environmental risk factors and suicide attempts. The large, population-based sample permitted the opportunity to separate lesbians and gays from bisexual youth in the statistical analyses. The LGB and heterosexual participants were recruited using identical sampling methods. Many previous studies45
have recruited LGB respondents from different venues than heterosexuals, which may introduce sampling biases.46
An additional methodological strength is our objective measure of the social environment. Previous studies have used self-report measures of the environment, such as perceived discrimination.47
These subjective measures may capture how LGB individuals construe their experience of living in harmful social environments, but such measures are confounded with mental health status.15
In contrast, our index of the social environment occurred outside the control of the individual and could not be caused by individual-level factors that also might affect the dependent variable, which helps to minimize endogeneity. Finally, many studies using ecological data suffer from the “ecological fallacy” when they try to extrapolate from aggregated data to individuals.48
However, a strength of the current study was the assessment of suicide attempts on the individual level, thus linking the ecologic with the individual level and avoiding incorrect inference across levels.49