Link between number of romantic partners you have and substance abuse
February 14, 2013 9:21 PM Subscribe
The correlation discovered at Otago University between the number of romantic partners you have and substance abuse
"Changes in sexual behavior have resulted in longer periods of multiple serial or concurrent relationships. This study investigated the effects of multiple heterosexual partners on mental health, specifically, whether higher numbers of partners were linked to later anxiety, depression, and substance dependency. Data from the Dunedin Multidisciplinary Health and Development Study, a prospective, longitudinal study of a birth cohort born in 1972–1973 in Dunedin, New Zealand were used". [Paper] [Poster]
"... Increasing numbers of sex partners were associated with increasing risk of substance dependence disorder at all three ages. The association was stronger for women and remained after adjusting for prior disorder"
"Changes in sexual behavior have resulted in longer periods of multiple serial or concurrent relationships. This study investigated the effects of multiple heterosexual partners on mental health, specifically, whether higher numbers of partners were linked to later anxiety, depression, and substance dependency. Data from the Dunedin Multidisciplinary Health and Development Study, a prospective, longitudinal study of a birth cohort born in 1972–1973 in Dunedin, New Zealand were used". [Paper] [Poster]
"... Increasing numbers of sex partners were associated with increasing risk of substance dependence disorder at all three ages. The association was stronger for women and remained after adjusting for prior disorder"
For some reason, this article/study makes me really uncomfortable. Maybe because it comes off as sort of slut-shaming? I think maybe my reaction here is a little knee-jerk, but I'll be interested in the other responses people have to this one.
posted by SkylitDrawl at 9:32 PM on February 14, 2013 [7 favorites]
posted by SkylitDrawl at 9:32 PM on February 14, 2013 [7 favorites]
I would like to see a follow-up study about how rock & roll fits into the picture.
posted by aubilenon at 9:34 PM on February 14, 2013 [42 favorites]
posted by aubilenon at 9:34 PM on February 14, 2013 [42 favorites]
The study found that men born in Dunedin in 1972-73 were still more likely than women to have multiple sexual partners between 18 and 31.
I'm not sure that this is mathematically possible, unless many of the men were either having sex with other men or with women younger than 18 or older than 31.
posted by empath at 9:37 PM on February 14, 2013
I'm not sure that this is mathematically possible, unless many of the men were either having sex with other men or with women younger than 18 or older than 31.
posted by empath at 9:37 PM on February 14, 2013
It sounds like they found an association between number of sexual partners and substance dependence disorder, and it was a closer association in women than men. It's interesting, but it raises more questions than it answers (which can be a good thing, of course); we did not learn anything from the study about why that association exists, which seems like the central question to me.
posted by clockzero at 9:40 PM on February 14, 2013 [1 favorite]
posted by clockzero at 9:40 PM on February 14, 2013 [1 favorite]
A news article about a single study of a cohort of unknown size published in an obscure journal behind a pay wall with charged conclusions? Flagged as not the best of the web.
posted by Slarty Bartfast at 9:42 PM on February 14, 2013 [6 favorites]
posted by Slarty Bartfast at 9:42 PM on February 14, 2013 [6 favorites]
For some reason, this article/study makes me really uncomfortable. Maybe because it comes off as sort of slut-shaming?
This definitely came off as subtly slut-shaming to me:
One of the Dunedin authors, public health professor Charlotte Paul, said the study pointed to mental health risks for women in having many sexual partners, and the obvious physical risks of sexually transmitted diseases.
"At present we are engaged in some very proper discussion about individuals committing to each other in marriage for gay people, but we don't seem to be talking about this at all in terms of heterosexual activity."
Also, it seems like some of those quoted (or maybe just the author of the article) were leaping to some very unsupported conclusion about correlation vs. causality.
posted by lunasol at 9:42 PM on February 14, 2013 [2 favorites]
This definitely came off as subtly slut-shaming to me:
One of the Dunedin authors, public health professor Charlotte Paul, said the study pointed to mental health risks for women in having many sexual partners, and the obvious physical risks of sexually transmitted diseases.
"At present we are engaged in some very proper discussion about individuals committing to each other in marriage for gay people, but we don't seem to be talking about this at all in terms of heterosexual activity."
Also, it seems like some of those quoted (or maybe just the author of the article) were leaping to some very unsupported conclusion about correlation vs. causality.
posted by lunasol at 9:42 PM on February 14, 2013 [2 favorites]
The thing that was really clanging all the slut-shaming bells for me was the title of the news article: Women with more sex partners turn to drink and drugs. It is like the article is implying that women with multiple sexual partners are bad or mentally unhealthy people because they enjoy having sex. Maybe I am reading way too much into this, but it just squicked me out.
posted by SkylitDrawl at 9:48 PM on February 14, 2013 [1 favorite]
posted by SkylitDrawl at 9:48 PM on February 14, 2013 [1 favorite]
Please repeat as often as necessary (which in this case, should be continuously until you've forgotten ever reading the linked article):
CORRELATION DOES NOT INFER CAUSATION.
CORRELATION DOES NOT INFER CAUSATION.
CORRELATION DOES NOT INFER CAUSATION.
posted by oneswellfoop at 9:51 PM on February 14, 2013 [8 favorites]
CORRELATION DOES NOT INFER CAUSATION.
CORRELATION DOES NOT INFER CAUSATION.
CORRELATION DOES NOT INFER CAUSATION.
posted by oneswellfoop at 9:51 PM on February 14, 2013 [8 favorites]
I'm not sure that this is mathematically possible, unless many of the men were either having sex with other men or with women younger than 18 or older than 31.
You're right that, ignoring gay and outside-age-range pairings, the mean number of sexual partners should be the same for men and women. However, the distribution of partner-counts can be different, leading to different medians or other statistical measures. It depends on what precisely they mean by "more likely". As an extreme example, suppose that only two women have any partners, and the rest have zero, and all the men partner with those two women. Then all the men would have two partners, while almost none of the women have any, so in that situation a man would be more likely to have multiple partners than a woman (since 100% of the men would have 1+ partners, while ~0% of the women would). Likewise, the median number of partners for men will be two and for women zero, although the mean will be two for both sides.
posted by Pyry at 9:52 PM on February 14, 2013 [12 favorites]
You're right that, ignoring gay and outside-age-range pairings, the mean number of sexual partners should be the same for men and women. However, the distribution of partner-counts can be different, leading to different medians or other statistical measures. It depends on what precisely they mean by "more likely". As an extreme example, suppose that only two women have any partners, and the rest have zero, and all the men partner with those two women. Then all the men would have two partners, while almost none of the women have any, so in that situation a man would be more likely to have multiple partners than a woman (since 100% of the men would have 1+ partners, while ~0% of the women would). Likewise, the median number of partners for men will be two and for women zero, although the mean will be two for both sides.
posted by Pyry at 9:52 PM on February 14, 2013 [12 favorites]
Please repeat as often as necessary
That never needs to repeated on Metafilter, ever.
posted by smoke at 10:14 PM on February 14, 2013 [12 favorites]
That never needs to repeated on Metafilter, ever.
posted by smoke at 10:14 PM on February 14, 2013 [12 favorites]
This doesn't really seem that surprising to me. Someone who is drawn to "thrill seeking" type behavior is pretty clearly more likely to have more sex and do more drugs than someone who strives for traditional propriety.
posted by aubilenon at 10:21 PM on February 14, 2013 [7 favorites]
posted by aubilenon at 10:21 PM on February 14, 2013 [7 favorites]
empath: "I'm not sure that this is mathematically possible, unless many of the men were either having sex with other men or with women younger than 18 or older than 31."
Or with women not born in Dunedin. I'm assuming the city's not under an impregnable dome.
posted by Joakim Ziegler at 10:29 PM on February 14, 2013
Or with women not born in Dunedin. I'm assuming the city's not under an impregnable dome.
posted by Joakim Ziegler at 10:29 PM on February 14, 2013
PEOPLE WHO LIKE TO DO SOME THINGS THAT ARE FUN, SUCH AS DRUGS, ARE ALSO LIKELY TO DO OTHER THINGS THAT ARE FUN, SUCH AS HAVING SEX.
posted by dunkadunc at 10:34 PM on February 14, 2013 [6 favorites]
posted by dunkadunc at 10:34 PM on February 14, 2013 [6 favorites]
PEOPLE WHO LIKE TO DO SOME THINGS THAT ARE FUN, SUCH AS DRUGS, ARE ALSO LIKELY TO DO OTHER THINGS THAT ARE FUN, SUCH AS HAVING SEX.
The study counted instances of alcohol abuse and not casual drinking
From the abstract:
"This study established a strong association between number of sex partners and later substance disorder, especially for women, which persisted beyond prior substance use and mental health problems more generally."
posted by mataboy at 10:39 PM on February 14, 2013
The study counted instances of alcohol abuse and not casual drinking
From the abstract:
"This study established a strong association between number of sex partners and later substance disorder, especially for women, which persisted beyond prior substance use and mental health problems more generally."
posted by mataboy at 10:39 PM on February 14, 2013
Okay. So what?
posted by SkylitDrawl at 10:47 PM on February 14, 2013
posted by SkylitDrawl at 10:47 PM on February 14, 2013
CORRELATION DOES NOT INFER CAUSATION.
CORRELATION DOES NOT INFER CAUSATION.
CORRELATION DOES NOT INFER CAUSATION.
posted by oneswellfoop at 9:51 PM on 2/14
You Keep Using That Word, I Do Not Think It Means What You Think It Means
posted by Doroteo Arango II at 11:12 PM on February 14, 2013 [21 favorites]
I'm not sure I believe the bit about the relationship between number of sexual partners and substance abuse being stronger for women than for men. As far as I can tell from reading the paper, the required interaction effect only shows up for the oldest age category, so the authors are basically requiring us to believe in an age x gender x number.of.partners three-way interaction. However, they don't actually report an explicit test for that interaction, which feels a little unconvincing to me.
The rest of it I believe, at least the part that refers to people with <2.5 partners per year. Not sure they have the sample size to extrapolate very much beyond that range. Admittedly, they do get some pretty large effect sizes even with the range truncated: the incidence of subsequent substance abuse doubles among people with 1-2.5 partners relative to people with 0-1 partners. For instance, it jumps from from 18.4% to 37.0% among men aged 18-20. That's kind of surprising to me.
posted by mixing at 11:17 PM on February 14, 2013
The rest of it I believe, at least the part that refers to people with <2.5 partners per year. Not sure they have the sample size to extrapolate very much beyond that range. Admittedly, they do get some pretty large effect sizes even with the range truncated: the incidence of subsequent substance abuse doubles among people with 1-2.5 partners relative to people with 0-1 partners. For instance, it jumps from from 18.4% to 37.0% among men aged 18-20. That's kind of surprising to me.
posted by mixing at 11:17 PM on February 14, 2013
I'm going to start hitting myself in the forehead with a mallet over the correlation != causation thing. It won't stop people from posting it but at least I won't care.
I'm not sure that this is mathematically possible, unless many of the men were either having sex with other men or with women younger than 18 or older than 31.
Pyry addressed this but I'll try to be more direct and concise. It is absolutely possible and all it requires is that a small number of women each have sex with a whole lotta men. So most women have fewer sexual partners than the average man but the women who have more sexual partners than the average man apply themselves very diligently to the process.
In other words, the mean for men and women would be the same but the standard deviation would be quite different.
posted by Justinian at 11:25 PM on February 14, 2013 [2 favorites]
I'm not sure that this is mathematically possible, unless many of the men were either having sex with other men or with women younger than 18 or older than 31.
Pyry addressed this but I'll try to be more direct and concise. It is absolutely possible and all it requires is that a small number of women each have sex with a whole lotta men. So most women have fewer sexual partners than the average man but the women who have more sexual partners than the average man apply themselves very diligently to the process.
In other words, the mean for men and women would be the same but the standard deviation would be quite different.
posted by Justinian at 11:25 PM on February 14, 2013 [2 favorites]
aubilenon: "This doesn't really seem that surprising to me. Someone who is drawn to "thrill seeking" type behavior is pretty clearly more likely to have more sex and do more drugs than someone who strives for traditional propriety."
That's also the first suggestion in the paper ("sexual risk taking and substance use may be part of the cluster of risk taking behaviors common in adolescence and young adulthood"). I think the surprising thing isn't so much that the effect exists, but that it's weirdly large. I'm a bit surprised that going from "one or less partner per year" to "one or two partners per year" should somehow be associated with such a large increase in the probability of substance abuse problems. It just doesn't feel like that big a difference in terms of sexual behaviour, but it's a pretty big difference in terms of substance abuse outcomes.
posted by mixing at 11:29 PM on February 14, 2013
That's also the first suggestion in the paper ("sexual risk taking and substance use may be part of the cluster of risk taking behaviors common in adolescence and young adulthood"). I think the surprising thing isn't so much that the effect exists, but that it's weirdly large. I'm a bit surprised that going from "one or less partner per year" to "one or two partners per year" should somehow be associated with such a large increase in the probability of substance abuse problems. It just doesn't feel like that big a difference in terms of sexual behaviour, but it's a pretty big difference in terms of substance abuse outcomes.
posted by mixing at 11:29 PM on February 14, 2013
Assuming that it's accurate, perhaps it can be explained this way: people with no partners are perhaps among those more likely to not be involved in alcohol-related socializing (no partying, clubbing), maybe because they are shy, or have social anxieties, or are just studying/working too much, and people with only one partner are more likely in a committed relationship, which often means less drug/alcohol-related social behavior?
posted by taz at 11:39 PM on February 14, 2013
posted by taz at 11:39 PM on February 14, 2013
I always drink a lot less when I'm in a relationship, because I don't go to bars and clubs as much.
posted by empath at 11:58 PM on February 14, 2013 [1 favorite]
posted by empath at 11:58 PM on February 14, 2013 [1 favorite]
What if the partners aren't romantic but more, like dirty/sexy/nasty?
posted by From Bklyn at 12:03 AM on February 15, 2013
posted by From Bklyn at 12:03 AM on February 15, 2013
taz: "Assuming that it's accurate, perhaps it can be explained this way: people with no partners are perhaps among those more likely to not be involved in alcohol-related socializing (no partying, clubbing), maybe because they are shy, or have social anxieties, or are just studying/working too much, and people with only one partner are more likely in a committed relationship, which often means less drug/alcohol-related social behavior?"
Yeah, that's pretty similar to what I was thinking, and to be fair to the authors they do suggest "shared social context" as one of the possible explanations for the effect. It would have been nice if the study had included some measure of social activity to include in the analysis in order to test the idea. But not every study can measure everything of interest, I suppose.
posted by mixing at 12:03 AM on February 15, 2013
Yeah, that's pretty similar to what I was thinking, and to be fair to the authors they do suggest "shared social context" as one of the possible explanations for the effect. It would have been nice if the study had included some measure of social activity to include in the analysis in order to test the idea. But not every study can measure everything of interest, I suppose.
posted by mixing at 12:03 AM on February 15, 2013
It is entirely possible that women who are mentally unwell tend to have more sex partners in order to feel valued, to compensate for negative self-image. Our society is not kind to young women who suddenly find themselves objects of sexual desire at a young age, especially if they display any emotional vulnerability or weak boundaries. Unfortunately, vulnerable women often get taken advantage of, and whatever problems they carry get heavier with time. Hence, self-medication with drugs or alcohol. I admit that I have no idea how common such scenarios are, but I am frankly a little surprised they didn't consider this possibility. The moralizing about 'risk-taking behavior' and the implication that people who choose more sex partners are opening themselves up to mental health and addiction problems seems a little narrow to me.
posted by PercussivePaul at 1:09 AM on February 15, 2013 [2 favorites]
posted by PercussivePaul at 1:09 AM on February 15, 2013 [2 favorites]
Perhaps the correlation is that there is a factor (novelty seeking? sensation seeking?) that increases the likelihood of one having more sexual partners and of one using drugs/alcohol?
posted by acb at 1:37 AM on February 15, 2013
posted by acb at 1:37 AM on February 15, 2013
I'm not sure this is about "slut shaming" and probably, more about dopamine receptors. ADHD is another version of variation in dopamine, dopamine receptors and transport, and the brain's reward system, also correlated with higher rates of addiction.
posted by C.A.S. at 3:09 AM on February 15, 2013 [1 favorite]
posted by C.A.S. at 3:09 AM on February 15, 2013 [1 favorite]
Can the title and link text be changed? The study was about sexual partners, not romance. The two are distinct, and thus it's misleading.
posted by explosion at 4:31 AM on February 15, 2013
posted by explosion at 4:31 AM on February 15, 2013
This study may be more about 1980s Dunedin culture than anything else.
posted by Obscure Reference at 4:33 AM on February 15, 2013 [1 favorite]
posted by Obscure Reference at 4:33 AM on February 15, 2013 [1 favorite]
Well, I admit I find it annoying in a knee-jerk sort of way, because a) the perverse and obsessive fascination and judgement related to anything about girls/women and sex, plus b) the impossibility of anyone or anything existing in the glare of all this hyperattention (I don't say "observation" because I feel it falls far short of benign observation, which in itself would create odd patterns) without some pretty convoluted effects. So the number of sex partners in relation to substance abuse is far from a chicken or egg hypothetical to me, because in either case there has been so, so, so much more that happened before that, beginning at birth, arguably.
I can think of a lot of different, wildly varying, reasons that a woman may have both more sexual partners and a substance abuse problem, wherein the latter doesn't arise from the former, and/or wherein both are behaviors that arise from something else (or several other things) entirely. Maybe this is yet another product of reportage failure, but this "news" once again has me feeling like gently explaining that women are actually more complex than a single-cell amoeba.
Oh, wow! We thought some women had more sex partners because they got wasted, but now we learn that women get wasted because they have more sex partners!! Verrrrry interrrrrresting. *steeples hands, furrows brow* How silly.
posted by taz at 4:34 AM on February 15, 2013
I can think of a lot of different, wildly varying, reasons that a woman may have both more sexual partners and a substance abuse problem, wherein the latter doesn't arise from the former, and/or wherein both are behaviors that arise from something else (or several other things) entirely. Maybe this is yet another product of reportage failure, but this "news" once again has me feeling like gently explaining that women are actually more complex than a single-cell amoeba.
Oh, wow! We thought some women had more sex partners because they got wasted, but now we learn that women get wasted because they have more sex partners!! Verrrrry interrrrrresting. *steeples hands, furrows brow* How silly.
posted by taz at 4:34 AM on February 15, 2013
Slarty Bartfast: A news article about a single study of a cohort of unknown size published in an obscure journal behind a pay wall with charged conclusions? Flagged as not the best of the web.You know, it's actually not too hard to find out information about the Dunedin Multidisciplinary Health and Development Study. Just because it's taking place in a funny little country on the other side of the world doesn't (necessarily) mean that it's unscientific, obscure, or wrong.
posted by Sonny Jim at 4:41 AM on February 15, 2013 [4 favorites]
> The study found that men born in Dunedin in 1972-73 were still more likely than women to have multiple sexual partners between 18 and 31.
>>I'm not sure that this is mathematically possible
Do Men and Women Report Their Sexual Partnerships Differently? Evidence from Kisumu, Kenya
posted by flug at 5:46 AM on February 15, 2013
>>I'm not sure that this is mathematically possible
Do Men and Women Report Their Sexual Partnerships Differently? Evidence from Kisumu, Kenya
they found that, on average, women reported significantly fewer nonmarital sexual partners than men, as well as longer sexual relationships, leading them to conclude that in general men "swagger" (i.e., exaggerate their sexual activities), while women are "secretive" (i.e., underreport their sexual behaviors).There's more to the story, but the overreporting/underreporting thing seems to be a pretty large effect.
posted by flug at 5:46 AM on February 15, 2013
Man, the Dunedin are really struggling to justify their budget after the unmaking of the Ring, eh?
posted by snuffleupagus at 6:01 AM on February 15, 2013
posted by snuffleupagus at 6:01 AM on February 15, 2013
This study may be more about 1980s Dunedin culture than anything else.
It's all the awesome music they were listening to.
/born in Dunedin, but two years post-cohort births.
posted by gaspode at 6:01 AM on February 15, 2013 [1 favorite]
It's all the awesome music they were listening to.
/born in Dunedin, but two years post-cohort births.
posted by gaspode at 6:01 AM on February 15, 2013 [1 favorite]
For some reason, this article/study makes me really uncomfortable. Maybe because it comes off as sort of slut-shaming? I think maybe my reaction here is a little knee-jerk, but I'll be interested in the other responses people have to this one.
Well,obviously one does not know why these things correlate - but if you want a totally speculative potential explanation that does not slut-shame, what about this:
We live in a society which is structured around families and couples such that if you're having lots of partners it can be hard to build intimacy - that is, it's not like joyous poly relationships are the norm; we live in a society where there's strong messages about how marriage/long-term relationships and two-parent families are the norm and a sign of being a good, functioning adult; we live in a society where women in particular struggle financially if they are single parents and/or don't have a second income to rely on; and we live in a society which considers women who have a lot of partners to be damaged in some way.
Could any of that add up to "people who either really, really want lots of partners or who get into a life of less stable relationships face a lot of social messaging and material risk that encourages substance abuse"?
I tend to think that this kind of study is really useless, actually, and wish that the money that went into this sort of thing could be put into, say, a really good shelter program for women escaping abusive relationships. Or into community gardens. Or skateparks for bored kids. Or pretty much anything run at the local level that meets the actual, self-reported needs of individuals.
posted by Frowner at 6:09 AM on February 15, 2013
Well,obviously one does not know why these things correlate - but if you want a totally speculative potential explanation that does not slut-shame, what about this:
We live in a society which is structured around families and couples such that if you're having lots of partners it can be hard to build intimacy - that is, it's not like joyous poly relationships are the norm; we live in a society where there's strong messages about how marriage/long-term relationships and two-parent families are the norm and a sign of being a good, functioning adult; we live in a society where women in particular struggle financially if they are single parents and/or don't have a second income to rely on; and we live in a society which considers women who have a lot of partners to be damaged in some way.
Could any of that add up to "people who either really, really want lots of partners or who get into a life of less stable relationships face a lot of social messaging and material risk that encourages substance abuse"?
I tend to think that this kind of study is really useless, actually, and wish that the money that went into this sort of thing could be put into, say, a really good shelter program for women escaping abusive relationships. Or into community gardens. Or skateparks for bored kids. Or pretty much anything run at the local level that meets the actual, self-reported needs of individuals.
posted by Frowner at 6:09 AM on February 15, 2013
I would like to see a follow-up study about how rock & roll fits into the picture.I so want to do a gushy, follow-up FPP full of live footage featuring Shayne Carter and his amazing cheekbones, but I'm supposed to be working ...
posted by Sonny Jim at 6:11 AM on February 15, 2013 [1 favorite]
You just can't make a statement like this from any such correlation data:
"A researcher in another long-running study in Christchurch, Dr Joe Boden, said the Dunedin finding was "novel" because everyone until now had thought the drink/sex linkage ran only one way - from alcohol to sex."
I can't see how they could ever make such a casual directional statement in either direction from a study like this. There are to many factors that push together prudes and non-drinkers (religion, straightness, conservatism) that I find it hard to believe you could make any meaningful assertions about something like this.
posted by mary8nne at 6:13 AM on February 15, 2013
"A researcher in another long-running study in Christchurch, Dr Joe Boden, said the Dunedin finding was "novel" because everyone until now had thought the drink/sex linkage ran only one way - from alcohol to sex."
I can't see how they could ever make such a casual directional statement in either direction from a study like this. There are to many factors that push together prudes and non-drinkers (religion, straightness, conservatism) that I find it hard to believe you could make any meaningful assertions about something like this.
posted by mary8nne at 6:13 AM on February 15, 2013
This study investigated the effects of multiple heterosexual partners on mental health, specifically, whether higher numbers of partners were linked to later anxiety, depression, and substance dependency.
It's not just correlation; it's later correlation, which is I assume part of how they're drawing the correlation in the direction they are. I don't know for sure how strong that connection is, I haven't read the paper and I'm not really qualified to judge its merits, but things like correlation not equaling causation and confounding factors aren't like some Metafilter secret that people doing research are totally unaware of.
posted by Bulgaroktonos at 6:18 AM on February 15, 2013 [3 favorites]
It's not just correlation; it's later correlation, which is I assume part of how they're drawing the correlation in the direction they are. I don't know for sure how strong that connection is, I haven't read the paper and I'm not really qualified to judge its merits, but things like correlation not equaling causation and confounding factors aren't like some Metafilter secret that people doing research are totally unaware of.
posted by Bulgaroktonos at 6:18 AM on February 15, 2013 [3 favorites]
For some reason, this article/study makes me really uncomfortable. Maybe because it comes off as sort of slut-shaming? I think maybe my reaction here is a little knee-jerk, but I'll be interested in the other responses people have to this one.
Well, it seems to me that the thing that is missing is a comparison of other risk taking behaviors. Substance abuse, anxiety, etc are all known to be strongly correlated with poor impulse control and poor risk assessment.
Sex is risky, doing it safely is difficult - and so I think if you take poor sexual choices and look at them as just another risk-taking behavior the results shouldn't be so surprising. I bet the subjects also sped more, got into more fights and were generally more volatile overall.
So, yeah - the focus on the sex thing seems... patriarchal and sensationalist. It's just one way in which people can act out and I don't think it's really any better or worse than the other ways they do that.
posted by Pogo_Fuzzybutt at 6:49 AM on February 15, 2013 [1 favorite]
Well, it seems to me that the thing that is missing is a comparison of other risk taking behaviors. Substance abuse, anxiety, etc are all known to be strongly correlated with poor impulse control and poor risk assessment.
Sex is risky, doing it safely is difficult - and so I think if you take poor sexual choices and look at them as just another risk-taking behavior the results shouldn't be so surprising. I bet the subjects also sped more, got into more fights and were generally more volatile overall.
So, yeah - the focus on the sex thing seems... patriarchal and sensationalist. It's just one way in which people can act out and I don't think it's really any better or worse than the other ways they do that.
posted by Pogo_Fuzzybutt at 6:49 AM on February 15, 2013 [1 favorite]
I tend to think that this kind of study is really useless, actually, and wish that the money that went into this sort of thing could be put into, say, a really good shelter program for women escaping abusive relationships. Or into community gardens. Or skateparks for bored kids. Or pretty much anything run at the local level that meets the actual, self-reported needs of individuals.
Right, but this is a longitudinal cohort study. They've been studying this cohort since birth, looking at a bunch of variables. It's not specifically an investigation into sexual behavior and substance-related disorders. This is just the current data that's being reported. Everything is ongoing.
A selection of other papers published using data from the study (per the link to the study homepage)
:
Gingival condition and oral hygiene in 12 year old Dunedin children
Factors associated with doubled-up housing - a common precursor to homelessness
An assessment of motor ability in seven year olds
Intrauterine growth retardation and blood pressure at age seven and eighteen
No negative outcomes of childhood middle ear disease in adulthood
etc. etc.
posted by gaspode at 7:52 AM on February 15, 2013 [1 favorite]
Right, but this is a longitudinal cohort study. They've been studying this cohort since birth, looking at a bunch of variables. It's not specifically an investigation into sexual behavior and substance-related disorders. This is just the current data that's being reported. Everything is ongoing.
A selection of other papers published using data from the study (per the link to the study homepage)
:
Gingival condition and oral hygiene in 12 year old Dunedin children
Factors associated with doubled-up housing - a common precursor to homelessness
An assessment of motor ability in seven year olds
Intrauterine growth retardation and blood pressure at age seven and eighteen
No negative outcomes of childhood middle ear disease in adulthood
etc. etc.
posted by gaspode at 7:52 AM on February 15, 2013 [1 favorite]
Okay, as I mentioned above, almost all of the participants in the study reported having 2.5 sexual partners per year or fewer, so there are sample size problems at the upper end of the range. This problem is especially severe for women in the oldest age category: the estimated rate of substance abuse in this case is based on only 8 participants. Given that the only evidence that the paper provides for a supposed "gender difference" in the nature of the relationship between substance abuse and number of sexual partners comes from this group, I started worrying.
Having nothing better to do with my evening, I decided to pull as much of the raw data as I could from the paper, and reanalysed the data myself, restricting myself only to those people reporting 2.5 sexual partners or fewer, since that's where most of their data are. Because the authors only reported data aggregated into three groups, "0-1 partners", "1-2.5 partners", ">2.5 partners", I had to treat number of sexual partners as a binary variable: "0-1 partners" or "1-2.5 partners". With this restricted data set, the logistic regressions show significant effects of gender and number of sexual partners, but no significant interactions between the two, and certainly nothing like the supposed three way interaction that the paper implies. In other words, all I can find evidence for in this data set is the fact that men have more substance abuse problems than women (something that is pretty well-known already, I believe), and people with 1-2.5 sexual partners per year have more substance abuse problems than people with 0-1 partners. The nature of this relationship seemed to be the same for men as for women.
Given this, here's how I see the study. The central finding is that having more sexual partners is associated with a higher rate of subsequent substance abuse problems: this finding is really easy to reproduce, even if you reanalyse the data in a more conservative way like I did. The reasons for this finding are not clear, of course, and it's a pity that the study can't shed more light on why this effect occurs. But that's not the big problem here, as I see it. The big problem is that this supposed "it's different for women" bit is the least compelling part of the study, and yet for some reason this is the main focus of the newspaper article. I honestly can't think of a non-sexist explanation for that.
posted by mixing at 7:55 AM on February 15, 2013 [8 favorites]
Having nothing better to do with my evening, I decided to pull as much of the raw data as I could from the paper, and reanalysed the data myself, restricting myself only to those people reporting 2.5 sexual partners or fewer, since that's where most of their data are. Because the authors only reported data aggregated into three groups, "0-1 partners", "1-2.5 partners", ">2.5 partners", I had to treat number of sexual partners as a binary variable: "0-1 partners" or "1-2.5 partners". With this restricted data set, the logistic regressions show significant effects of gender and number of sexual partners, but no significant interactions between the two, and certainly nothing like the supposed three way interaction that the paper implies. In other words, all I can find evidence for in this data set is the fact that men have more substance abuse problems than women (something that is pretty well-known already, I believe), and people with 1-2.5 sexual partners per year have more substance abuse problems than people with 0-1 partners. The nature of this relationship seemed to be the same for men as for women.
Given this, here's how I see the study. The central finding is that having more sexual partners is associated with a higher rate of subsequent substance abuse problems: this finding is really easy to reproduce, even if you reanalyse the data in a more conservative way like I did. The reasons for this finding are not clear, of course, and it's a pity that the study can't shed more light on why this effect occurs. But that's not the big problem here, as I see it. The big problem is that this supposed "it's different for women" bit is the least compelling part of the study, and yet for some reason this is the main focus of the newspaper article. I honestly can't think of a non-sexist explanation for that.
posted by mixing at 7:55 AM on February 15, 2013 [8 favorites]
It's called the Novelty Seeking personality trait in psychology. This isn't really a surprising finding at all. In fact, I would imagine it's pretty well known that people who enjoy drugs also enjoy lots of sex.
Anyway the problem with saying this is that it's not really true exactly, and it sort of becomes a koan that people repeat whenever they want to ignore some science. You could easily see global warming denialist arguing that you can't imply causation from correlation.
In reality if A and B are correlated, you can actually infer that one of three things is true: A causes B, B causes A, or some third thing causes both. You can't tell which is true, but in this case, it's probably the case that having a novelty seeking personality would be the cause of both. There's also perhaps the self-restraint issue in play as well, since they looked at people with substance abuse issues, and not just people who reported using drugs.
On the other hand, if you know when things happen, you can rule things out. Things in the future can't go back in time and cause things to happen in the past, so if the people in the study were having lots of sex before they started the drugs, then you know the drugs didn't cause the sex.
posted by delmoi at 8:29 AM on February 15, 2013
I'm not sure that this is mathematically possible, unless many of the men were either having sex with other men or with women younger than 18 or older than 31.Or if all of the men sleep with a smaller subset of women - so for example, if you had 100 women and 100 men, if 66 of the women slept with only one man, and 34 of the women slept with 4 men each then every single man could have the opportunity to sleep with at least two women (and two lucky guys get to sleep with 3). The math is pretty simple.
CORRELATION DOES NOT INFER CAUSATION.Usually people say "imply" - "infer" would mean that the correlation itself was figuring out causation, the correct way to say that would be "you can't infer causation from correlation"
Anyway the problem with saying this is that it's not really true exactly, and it sort of becomes a koan that people repeat whenever they want to ignore some science. You could easily see global warming denialist arguing that you can't imply causation from correlation.
In reality if A and B are correlated, you can actually infer that one of three things is true: A causes B, B causes A, or some third thing causes both. You can't tell which is true, but in this case, it's probably the case that having a novelty seeking personality would be the cause of both. There's also perhaps the self-restraint issue in play as well, since they looked at people with substance abuse issues, and not just people who reported using drugs.
On the other hand, if you know when things happen, you can rule things out. Things in the future can't go back in time and cause things to happen in the past, so if the people in the study were having lots of sex before they started the drugs, then you know the drugs didn't cause the sex.
posted by delmoi at 8:29 AM on February 15, 2013
The article is pretty brief, and fits my own personal bias in relation to my observations of peers and acquaintances. I wasn't able to find the actual data through the link (computer challenged) but my ability to assess for accuracy is limited anyway so hearing others perspectives on that is more interesting.
My personal experience with women and men who have had a large number of sexual partners is that they do tend to be heavy drinkers. More likely to drink every time they socialize, more heavily when they drink and socialize, and more likely to drink on a daily basis.
I think for women, in order to put yourself in a vulnerable spot with new partners (which is more risky for women than men) alcohol is a pretty important aspect of hook-up culture and casual sex culture. And in order to forget and ignore that you might feel sad after non-commital sex, which puts a damper on being in the mood. Alcohol helps over come that aspect of non-commital sex. I kind of wonder if... time spent in hook-up culture... correlates with worse mental health. I also wonder if time spent binge drinking can itself increase risk of people with "non-alcoholic" genetic types of falling into abuse and dependance problems. Meaning that the more time you spend binge drinking in order to enjoy casual sex, the more likely you are to have some problems with alcohol later on.
These are just my wondering. Anacdata is much less interesting than data, but all this article and the discussion about the actual data due for me is make me thirst for larger quantities of data and hashing out of the variables involved in this apparent association.
It's a huge issue for me, because personally informed consent is pretty important for me and if people don't have accurate information about the risks they are taking I consider that a problem. I think we have done a wonderful thing by opening up sexual norms to individual choice (and including women in that) but that doesn't actually mean that human beings are all that skilled at understanding how their choices will affect them in the long run. I do think some social mores actually DOn sometimes serve protective functions for people who'se impulses don't line up with accurate assessment of the risks to themselves or people around them. The key is to make those mores actual fit reality rather than superstition (or misogyny, or prejudice etc).
Research into the effects of human behaviors is extremely useful as far as I'm concerned (but brings up important questions about how to deal with people who'se behaviors are counter their own needs or who are unable to understand or believe the data). I think people are slow to react to new data and that's because "firm" research conclusions are released all the time and frequently taken back when found totally wrong 10 years later. There's a certain amount of understandable "What the fuck to these researchers know, I just know this feels good and seems to suit me right now" that I think we all go through when research of the time doesn't add up to what we want to be true in the moment. It's really the same response people have to parental control as well. And hey, sometimes parents, and researchers, really ARE wrong, so having people willing to take risks and find out what the ACTUAL effects are is a great thing. (But bad for people who actually deal with crappy consequences from their risk taking). Measuring those results and using them to help others make accurate assessments of levels of risk/enjoyment in the future is a great thing that could result from that.
posted by xarnop at 8:31 AM on February 15, 2013
My personal experience with women and men who have had a large number of sexual partners is that they do tend to be heavy drinkers. More likely to drink every time they socialize, more heavily when they drink and socialize, and more likely to drink on a daily basis.
I think for women, in order to put yourself in a vulnerable spot with new partners (which is more risky for women than men) alcohol is a pretty important aspect of hook-up culture and casual sex culture. And in order to forget and ignore that you might feel sad after non-commital sex, which puts a damper on being in the mood. Alcohol helps over come that aspect of non-commital sex. I kind of wonder if... time spent in hook-up culture... correlates with worse mental health. I also wonder if time spent binge drinking can itself increase risk of people with "non-alcoholic" genetic types of falling into abuse and dependance problems. Meaning that the more time you spend binge drinking in order to enjoy casual sex, the more likely you are to have some problems with alcohol later on.
These are just my wondering. Anacdata is much less interesting than data, but all this article and the discussion about the actual data due for me is make me thirst for larger quantities of data and hashing out of the variables involved in this apparent association.
It's a huge issue for me, because personally informed consent is pretty important for me and if people don't have accurate information about the risks they are taking I consider that a problem. I think we have done a wonderful thing by opening up sexual norms to individual choice (and including women in that) but that doesn't actually mean that human beings are all that skilled at understanding how their choices will affect them in the long run. I do think some social mores actually DOn sometimes serve protective functions for people who'se impulses don't line up with accurate assessment of the risks to themselves or people around them. The key is to make those mores actual fit reality rather than superstition (or misogyny, or prejudice etc).
Research into the effects of human behaviors is extremely useful as far as I'm concerned (but brings up important questions about how to deal with people who'se behaviors are counter their own needs or who are unable to understand or believe the data). I think people are slow to react to new data and that's because "firm" research conclusions are released all the time and frequently taken back when found totally wrong 10 years later. There's a certain amount of understandable "What the fuck to these researchers know, I just know this feels good and seems to suit me right now" that I think we all go through when research of the time doesn't add up to what we want to be true in the moment. It's really the same response people have to parental control as well. And hey, sometimes parents, and researchers, really ARE wrong, so having people willing to take risks and find out what the ACTUAL effects are is a great thing. (But bad for people who actually deal with crappy consequences from their risk taking). Measuring those results and using them to help others make accurate assessments of levels of risk/enjoyment in the future is a great thing that could result from that.
posted by xarnop at 8:31 AM on February 15, 2013
My personal anecdata suggests that it's not implausible. A close friend fell off the wagon partly because he was sexually involved with small-time dealers (who apparently were not monogamous, or all that great of boyfriends). When he turned around and started producing and dealing himself, he was sheepishly honest about trading drugs for sex. So it's my impression that at least with in that particular sub-culture, that sex for drugs and sex with drugs were a thing.
I should probably say that I didn't witness the stereotype of the evil dealer who uses drugs as a lever for sexual coercion (although some of my friend's boyfriends came close). It was more of a casual, "let's get high together and fuck" sort of dynamic. And for him, getting high with people and having sex with people were psychologically related as well.
Granted, in many ways he was an unusual case. But if what xarnop's post suggests is true, it doesn't take many people like him in a small dataset to bust the curve.
posted by CBrachyrhynchos at 8:50 AM on February 15, 2013
I should probably say that I didn't witness the stereotype of the evil dealer who uses drugs as a lever for sexual coercion (although some of my friend's boyfriends came close). It was more of a casual, "let's get high together and fuck" sort of dynamic. And for him, getting high with people and having sex with people were psychologically related as well.
Granted, in many ways he was an unusual case. But if what xarnop's post suggests is true, it doesn't take many people like him in a small dataset to bust the curve.
posted by CBrachyrhynchos at 8:50 AM on February 15, 2013
The big problem is that this supposed "it's different for women" bit is the least compelling part of the study, and yet for some reason this is the main focus of the newspaper article. I honestly can't think of a non-sexist explanation for that.
This is totally what Taz was talking about, I think. Your analysis is fascinating, and I wish to hell we could be (as a society, not this thread specifically) talking about things in a much less gender-difference-focused sort of way, because there's more value in talking about social patterns around alcohol abuse than "sluts are drunk and crazy, amirite?"
posted by restless_nomad at 8:57 AM on February 15, 2013
This is totally what Taz was talking about, I think. Your analysis is fascinating, and I wish to hell we could be (as a society, not this thread specifically) talking about things in a much less gender-difference-focused sort of way, because there's more value in talking about social patterns around alcohol abuse than "sluts are drunk and crazy, amirite?"
posted by restless_nomad at 8:57 AM on February 15, 2013
Yeah it makes me EXTREMELY curious to know lot's of other data, such as what was the nature of the relationships with the partners? What was the relationship of alcohol use IN the sexual encounters/relationships? Were the partners emotionally available or indifferent to each other outside of the sex? How was this effect mediated by levels of commitment and emotional availability within the relationships? Were the relationships intended to lead toward potential marriage, or intended to be time limited or non-commital? What were the expectations on the women and men and how happy with the relationship set ups where the women and men while they were happening? How hopeful did the women and men feel that the relationship style they really wanted would be available in the opposite sex to begin and how much were they assuming compromise from there strongest desires was innately necessary for the sake of any relationship at all? I think there are a lot of factors in sexual relationships that could cause serious harm to people, but figuring out what those are includes so many variables, including the unique needs of indivuals themselves that it's pretty hard to create a standardized: "All people are harmed by unloving sex" or "All people are harmed by serial monogamy" or "All people are harmed by stuffing sexuality into only sex during marriage to one lifelong partner"
I think there are people who could be harmed by all of these things, and people who could be benefited by all of these things. Human needs vary from person to person, situation to situation. Still general trends might be helpful in assessing how risky a potential set up might be to, statistically speaking, leading to future ill effects for yourself. It might be helpful to know that people who turn out to be benefited in the long run by a long string of non-commital and emotionally distant partners tend to be one in a hundred verses 3 out of 5. And what percent THINK they will be benefitted by this vs actually turn out to be benefited (I.e. how accurate are humans usually at assesing how we will be affected by different kinds of set ups in the long run?)
Knowing how well the average person is able to predict future effects of their current desired relationship style might help individuals decide how much stock to place in their "instincts" about how they will be effected over the statistically likely effects demonstrated repeatedly in research.
(Personally I'm not just talking about sex here but many decisions. I think humans are very limited at predicted long term emotional and psychological effects of various decisions they make. Knowing the ins and outs of that as measured with science could probably spare a lot of people a lot of decisions that aren't likely to pan out well for them.)
posted by xarnop at 9:12 AM on February 15, 2013
I think there are people who could be harmed by all of these things, and people who could be benefited by all of these things. Human needs vary from person to person, situation to situation. Still general trends might be helpful in assessing how risky a potential set up might be to, statistically speaking, leading to future ill effects for yourself. It might be helpful to know that people who turn out to be benefited in the long run by a long string of non-commital and emotionally distant partners tend to be one in a hundred verses 3 out of 5. And what percent THINK they will be benefitted by this vs actually turn out to be benefited (I.e. how accurate are humans usually at assesing how we will be affected by different kinds of set ups in the long run?)
Knowing how well the average person is able to predict future effects of their current desired relationship style might help individuals decide how much stock to place in their "instincts" about how they will be effected over the statistically likely effects demonstrated repeatedly in research.
(Personally I'm not just talking about sex here but many decisions. I think humans are very limited at predicted long term emotional and psychological effects of various decisions they make. Knowing the ins and outs of that as measured with science could probably spare a lot of people a lot of decisions that aren't likely to pan out well for them.)
posted by xarnop at 9:12 AM on February 15, 2013
Also, possibly people who are prone to addiction and anxiety might make worse partners, leading to more failed relationships that they didn't want to lose. Which is depressing and leads to worse mental health. Sorry, just keep wondering at the potential variables here. This study seems more like a tease and not very informative.... lol.
posted by xarnop at 9:15 AM on February 15, 2013
posted by xarnop at 9:15 AM on February 15, 2013
(That said, thanks for posting it, I'm always in favor of research even if it's main result is to inspire more research to get a better picture of what's happening.)
posted by xarnop at 9:19 AM on February 15, 2013
posted by xarnop at 9:19 AM on February 15, 2013
How was it determined that these people had "substance abuse disorder?" I can't access the data, but this is either self-reported or by diagnosis rate. That could leave out a whole lot of people who are sitting at home numbly drinking themselves to sleep every night.
posted by desuetude at 11:47 AM on February 15, 2013
posted by desuetude at 11:47 AM on February 15, 2013
« Older Carmageddon! | BUNK - a rock video for Nardwuar + the Evaporators Newer »
This thread has been archived and is closed to new comments
Nah. A lack of sex leads to a lot of alcohol, too. Trust me.
posted by Capt. Renault at 9:31 PM on February 14, 2013 [7 favorites]