The logic of this study is typical of many modern statistical studies:
1. the authors fit a certain model, adjusted for certain covariates, and observe that a certain effect of interest is statistically significant.
2. therefore, if (a) the model is "correct" i.e. sufficiently well approximates the observed data distribution, and (b) the regression coefficient and structural coefficient coincide i.e. the true generative DAG is compatible with the chosen set of covariates, then a causal association from the predictor to the outcome exists with high probability.
3. the authors simply assume (2) and therefore imply that a causal relation likely exists.
I expect sexual interest is just a weak proxy for general health - sexual function(including libido) relies on cardiovascular, endothelial, autonomic function, is impaired by under/over-weight, dysglycemia, sleep disordered breathing, poor anaerobic fitness.
Step 2. Solicit funds and grants to start programs designed to get Japanese men over 40 interested in sex.
(Europe and the United States immediately follow suit with programs for their men, meanwhile China organizes a yearly parade of its sexual supermen, Russia creates single Pornhub VIP account for the entire country and publishes the password in state newspapers.)
If you say 'a correlation exists between X and Y' yes. If you say 'X is linked to Y' most people will assume they're positively correlated, in line with the positive sentence structure. I too drew the opposite inference from the HN headline to the actual result of the study.
This is a neat, large study, but its a classic chicken and egg problem. Do you lack sexual interest because you are unhealthy, or does being unhealthy also make you not interested in sex. If you have heart problems or dying of cancer, you probably aren't looking to hook up all that much.
...but this was a historical look at 81 Korean imperial eunuchs from the 14th to early 20th centuries, and compared them to other male imperial court members.
Anyone want to signup for a modern test?
...but before you dismiss this completely - modern animal mammalian studies have confirmed this result and it also helps explain why women live longer. ...though the studies only looked at castration prior to puberty. That's important because overall height (a proxy for mass and cell count) is inversely correlated with longevity.
The headline should be redrafted; the study suggests that lack of interest in sex is correlated with higher rates of mortality. From the abstract: all-cause mortality was significantly higher among men who lacked sexual interest than men who had sexual interest.
Full title from the source: “Association between lack of sexual interest and all-cause mortality in a Japanese general population: The Yamagata prospective observational study”
I find the title here inline with that enough.
Further the link is a short but accurate summary of the paper. Seriously just read it and then comment.
Given that I quoted from the abstract, what makes you think I didn't read it? Multiple other people made the same observation about the rewritten headline.
1. the authors fit a certain model, adjusted for certain covariates, and observe that a certain effect of interest is statistically significant.
2. therefore, if (a) the model is "correct" i.e. sufficiently well approximates the observed data distribution, and (b) the regression coefficient and structural coefficient coincide i.e. the true generative DAG is compatible with the chosen set of covariates, then a causal association from the predictor to the outcome exists with high probability.
3. the authors simply assume (2) and therefore imply that a causal relation likely exists.