There is a really interesting paper titled "How readers understand causal and correlational expressions used in news headlines" (Adams, 2007). As you might gather from the title, it looks at how readers interpret news headlines, and in particular, how much causation they think the wording communicates.
Over the course of four controlled experiments, participants were shown short, headline-like claims that spanned a variety of subjects. After each headline, they were asked the question:
“According to the headline, to what extent does [X] cause [Y]?”
From there, they rated it on a 0–100 scale, from “definitely does not cause” to “definitely does cause.”
I was reminded of the study when I read a recent article in Sci News that used the phrase “tied to” to talk about the relationship between teen anxiety and depression and nicotine use.
While the 2007 study did not explicitly use “tied to”, it did test headlines with “linked to” and “associated with”.
When tested, respondents gave these phrases a 45-55 on the causality scale. The researchers suggested this should be categorised as “moderately causal”. In other words, even though “associated with” technically describes correlation, typical readers will read these headlines as something that could at least contribute to depression and anxiety among teens.
The studies
The article looks at three studies which come from the same University of Gothenburg research group.
They were:
“Gender-specific associations between tobacco and tobacco-free nicotine use and symptoms of anxiety and depression in Swedish adolescents.” - A cross-sectional questionnaire study of nearly 3,000 Swedish upper-secondary-school students.
“Sex-specific associations between tobacco use and youth mental health: prospective and cross-sectional findings from the LoRDIA Program.” - A longitudinal observational study following roughly 1,500 Swedish adolescents from lower secondary school into upper secondary school.
“Nicotine produces a sex- and subregion-specific hypoglutamatergic state in rodent amygdala which can be reversed by motor-skill training.” - A laboratory study in rats examining how repeated nicotine exposure affects glutamate signalling in different parts of the amygdala.
Taken together, the studies say that nicotine/tobacco use and poorer mental-health indicators occur together in Swedish adolescents. In particular, it seems this association is stronger amongst girls. However, crucially, the authors admit that “we still cannot determine cause and effect.”
The rat study is perhaps more novel. In female rats, repeated nicotine exposure changed how brain cells communicate in the basolateral amygdala, which is the part of the brain involved in processing emotions and stress.
The researchers speculate this change might make emotional regulation harder and could help explain why nicotine use can be difficult to stop. Crucially, the study did not show that the rats became anxious or depressed.
Which products had the worst association with anxiety and depression?
As mentioned above, the 2nd study looked at teenagers aged about 17 and compared those who had used cigarettes, Swedish snus, or both in the past year with those who had used neither. After the authors adjusted for sex, socioeconomic status and alcohol use, they found that.
A) Compared with no use, any past-year cigarette and/or snus use was associated with:
1.5x the odds of ever reporting depression
1.98x the odds of ever reporting suicide ideation.
B) Exclusive smoking was associated with:
1.61x depression
1.79x for suicide ideation.
C) Dual use of cigarettes & snus was associated with:
1.57x for depressive symptoms
2.41x for suicide ideation.
D) Exclusive Swedish snus use was associated with:
0.57x depression
1.21x for suicide ideation
What might jump out here is that exclusive Swedish snus use correlated with lower depression. Cue the snus cures depression headlines any day now.
Final thoughts
Nicotine’s association with mental health issues is complex. There are essentially three options here.
1) Teens with mental health difficulties are more likely to seek out and use nicotine because it might provide some pleasure, relief, distraction, etc.
2) Nicotine withdrawal or persistent changes in glutamate signalling can make teens restless, give a low mood, or cause anxiety-like symptoms.
3) Difficult domestic factors at home can increase the likelihood of both nicotine use and mental-health symptoms.
I think that number 2 is probably something that studies should be able to prove, if it were true. As always, correlation is not causation, and, for my money, 1 and 2 are more likely.
