Beware of the overly simplistic linking of social media use to negative mental health outcomes
A study by a prominent US psychologist has found a sharp rise in youth depression and suicide since the mid-2000s, and linked it to the rise of social media.
But before we start treating Facebook as an unmitigated health hazard, Professor Ian Hickie, co-director of the Brain and Mind Institute at the University of Sydney, says social media has many upsides for youth mental health that are being ignored amid a moral panic.
The study in the Journal of Abnormal Psychology is led by Professor Jean Twenge of San Diego State University, who specialises in intergenerational differences and the effects of technology on mental health. Her books include iGen: Why Today’s Super-Connected Kids Are Growing Up Less Rebellious, More Tolerant, Less Happy – and Completely Unprepared for Adulthood.
The study team took data from the US National Survey on Drug Use and Health and found the numbers who had experienced a major depressive episode in the past year had risen by 52% (8.7% to 13.2%) among 12-to-17-year-olds from 2005-17 and by 63% from 2009-17 (8.1% to 13.2%) among those aged 18-25.
Serious psychological distress in the past month was up by 71% in the 18-25s from 2008-17, and suicide-related outcomes (ideation, plans, attempts, and deaths) in the past year also increased. The increases were generally much weaker in those aged 26 and over.
The study used age-period-cohort statistical analysis to determine the effects were not attributable to age regardless of time period or vice versa, but to birth cohort – specifically, the “iGen” cohort born in the late 1990s.
Twenge et al. write that they “cannot entirely rule out increased likelihood to self-disclose among more recent cohorts” or the rise in opioid addiction as possible drivers of the findings.
But their preferred explanation is increased use of digital media, and along with it, changed patterns of social interaction, cyber-bullying, use of social media for information about suicide methods and exposure to images of suicide and self-injury.
They also posit loss of sleep associated with overuse of the internet and social media as a potential driver.
The authors conclude: “The results suggest a need for more research to understand the role of factors such as technology and digital media use and sleep disturbance may play in mood disorder and suicide-related outcomes, and to develop specialised interventions for younger cohorts.”
But Professor Hickie says the study has important limitations and is sceptical that the data even shows a real increase, let alone that the blame can be sheeted home to social media. He says reduced stigma and more reporting may account for apparent rises in depression and suicide.
“This remains controversial,” he tells The Medical Republic. “These things are not that easy to track over long periods of time. They’re talking about effects over 2005-17. But if you took the trends from the 1990s to now, you’d come to the completely opposite conclusion: that suicide has gone down since the invention of social media.”
Professor Hickie says an apparent increase in suicidal thoughts and behaviour has in fact been observed among teenagers, especially girls aged 12 to 15, over the past five years. However, this is confounded by the increase in people seeking care and reporting problems, and at an earlier age.
The fact that the Twenge et al. used survey data means everything in the study is affected by public attitudes – not excluding actual deaths by suicide.
“Even the willingness of coroners to report suicide in younger people is affected by secular trends,” he says. “There’s cases where people have clearly killed themselves but the coroner didn’t believe they had the intent, because they were so young. I’ve personally been involved in cases [where] families don’t want it reported as suicide.”
Even if the recent reversal of the downwards suicide trend is real, he says, the assumption that it’s due to social media is much more questionable.
“It’s the explanation du jour that social media accounts for everything. But the research around social media is much more mixed about the positives of social connectedness.
“At a population level I don’t think we know whether the positives outweigh the much-reported negatives.
“People connect with each other a lot and people who were excluded previously from a whole lot of normal social situations actually connect through social media: same-sex-attracted kids, kids with autism spectrum disorder, those with social anxiety and a range of other difficulties now can connect with others like them in ways they could never connect in small-world networks.
“There’s a lot of evidence of support for people with social difficulties, among them, those with suicidal behaviours. It’s not all negative. So I think the conclusion that social media is connected with negative outcomes is particularly simplistic.”
He says people are seeking help and interventions online, and that there exist many platforms for that purpose, including Australia’s ReachOut, which was one of the first in the world, and Project Synergy, which Professor Hickie is involved with.
“There’s a lot of anecdotal reports about bullying and sexting and pressure over body image. But there’s also the upside for all sorts of groups who were marginalised before social media and for ordinary teenagers whose everyday use increases their connectedness.
“In individual cases [of youth suicide] there may be a strong argument as to the role that particularly antisocial behaviour did play. But that skips all the positives that are clearly taking place at the same time.
“It’s an easy blame [for families of suicide victims]. Whether it was the straw that broke the camel’s back, whether it had a causative effect, much harder to know.”
Even the hypothesised loss of sleep due to social media is contestable, he says, since changed sleeping patterns are a well-known part of normal adolescent development.
“There’s a lot of debate in the sleep literature as to whether that’s changed or not,” he says. “People, out of genuine concern, have jumped from anecdote to conclusions that the data doesn’t really justify … I’m not inclined towards the moral panic view.”
He says, however, that without comprehensive surveillance data, any real cultural change would be slow to identify.
“Compared with other areas, like infectious diseases, we [in mental health] have very poor datasets. We could be doing a much better job.”
