Genetic correlations could reflect vertical pleitropy. Also in those Zietsch et al studies, aren't the phenotypic (purely behavioral ones say) between sexuality and the mental outcome very small? So even if genetics is explaining all of those otherwise small phenotypic correlations. Not sure how strong evidence genetic correlations are against minority stress mechanisms (that is, do they really rule out purely "psychosocial" minority stress mechanisms since doesn't minority stress theory predict a genetic corelations anyways? Not dispositive evidence either way). Interesting debate though, thanks.
It looks like I missed a few papers and there also some new ones out. Thanks, reviewing these will be worth a new post on its own! I think the mental health disparities are quite big, for example non-heterosexuals reported lifetime depression almost twice as often in one of the Zietsch papers. You are right that a genetic correlation is compatible with the minority stress hypothesis. If being gay is heritable and it increases your chances of being depressed through discrimination, the gayness-depression correlation should be about as genetic as gayness itself. See here:
for a more detailed version of this idea. The total absence of a nonshared-environmental correlation - that is, no MZ difference - is a falsification of the minority stress hypothesis, however. If it is discrimination that makes gay people depressed, you should be more depressed than your MZ twin brother if you are gay but he isn't. If we see this pattern, we can still invoke other explanations than minority stress, but if we don't, that is a big problem for this theory.
Agreed. Thanks for those further references. One way around the problem of minority stress theory predicting genetic correlations (even if the authors of minority stress don't specify this, which they should have really) is to have a better causal model perhaps (like the ones in your main post about the theory), and then use some kind of cross-lagged longitudinal and yet genetically sensitive model to test it. This also helps reduce any reverse causation. But almost no datasets like this.
Not unreasonable at all (although what you imply sounds more like that it is trauma that causes both psychiatric problems and homosexuality, rather than the former causes the latter). The problem is that whenever you see this pattern that a certain type of people come from certain types from families it is more often than not genetic confounding, which is why these twin controls are so important.
A possible explanation for these findings is that being a sexual minority is indeed causal for mental health problems, but not through discrimination, but just through cultural experiences. It must be hard on its own to experience that you are just not interested in the kind of relationship building most people do (and which can result in having a family) but it is compounded by the influences of the gay subculture, which promotes a hedonistic lifestyle that can't be great for your mental health. I can't say though what percentage of sexual minorities actually participate in this subculture.
Three further discordant MZ studies of sexual orientation and mental health outcomes here:
https://www.cambridge.org/core/journals/psychological-medicine/article/minority-stressors-rumination-and-psychological-distress-in-monozygotic-twins-discordant-for-sexual-minority-status/DF3DD4762FF83D7B26C36AE325279020
https://ajph.aphapublications.org/doi/full/10.2105/AJPH.2016.303573
https://www.cambridge.org/core/journals/psychological-medicine/article/psychiatric-morbidity-associated-with-samesex-sexual-behaviour-influence-of-minority-stress-and-familial-factors/72656B3B344FE385B58B741CFA0DCD59
Also this sibling-control design study:
https://link.springer.com/article/10.1007/s10654-018-0411-y
Finally, these studies also rule out correlated genetic factors using causal modelling in twin cohorts, so a powerful design:
https://link.springer.com/article/10.1007/s10508-022-02455-9
https://link.springer.com/article/10.1007/s10519-022-10130-x
Genetic correlations could reflect vertical pleitropy. Also in those Zietsch et al studies, aren't the phenotypic (purely behavioral ones say) between sexuality and the mental outcome very small? So even if genetics is explaining all of those otherwise small phenotypic correlations. Not sure how strong evidence genetic correlations are against minority stress mechanisms (that is, do they really rule out purely "psychosocial" minority stress mechanisms since doesn't minority stress theory predict a genetic corelations anyways? Not dispositive evidence either way). Interesting debate though, thanks.
It looks like I missed a few papers and there also some new ones out. Thanks, reviewing these will be worth a new post on its own! I think the mental health disparities are quite big, for example non-heterosexuals reported lifetime depression almost twice as often in one of the Zietsch papers. You are right that a genetic correlation is compatible with the minority stress hypothesis. If being gay is heritable and it increases your chances of being depressed through discrimination, the gayness-depression correlation should be about as genetic as gayness itself. See here:
https://www.cambridge.org/core/books/abs/handbook-of-research-methods-in-social-and-personality-psychology/behavior-genetic-research-methods/E9010347A247AA4BE3CB864F26A10158
https://www.annualreviews.org/doi/10.1146/annurev-psych-113011-143752
for a more detailed version of this idea. The total absence of a nonshared-environmental correlation - that is, no MZ difference - is a falsification of the minority stress hypothesis, however. If it is discrimination that makes gay people depressed, you should be more depressed than your MZ twin brother if you are gay but he isn't. If we see this pattern, we can still invoke other explanations than minority stress, but if we don't, that is a big problem for this theory.
Agreed. Thanks for those further references. One way around the problem of minority stress theory predicting genetic correlations (even if the authors of minority stress don't specify this, which they should have really) is to have a better causal model perhaps (like the ones in your main post about the theory), and then use some kind of cross-lagged longitudinal and yet genetically sensitive model to test it. This also helps reduce any reverse causation. But almost no datasets like this.
Not unreasonable at all (although what you imply sounds more like that it is trauma that causes both psychiatric problems and homosexuality, rather than the former causes the latter). The problem is that whenever you see this pattern that a certain type of people come from certain types from families it is more often than not genetic confounding, which is why these twin controls are so important.
A possible explanation for these findings is that being a sexual minority is indeed causal for mental health problems, but not through discrimination, but just through cultural experiences. It must be hard on its own to experience that you are just not interested in the kind of relationship building most people do (and which can result in having a family) but it is compounded by the influences of the gay subculture, which promotes a hedonistic lifestyle that can't be great for your mental health. I can't say though what percentage of sexual minorities actually participate in this subculture.