Tim Squirrell is a PhD candidate in Science and Technology Studies at the University of Edinburgh. His research focusses on construction and negotiation of authority and expertise on the internet, with a focus on fitness and nutrition communities.

Why are more young women being admitted to hospital for self-harming?

A recent story on the BBC says that figures from the NHS suggest that the number of girls (under 18) admitted to hospital has almost doubled in 2 decades, from 7,327 times in 1997 to 13,463 times in 2017.

On face, it’s an astounding statistic, and the media takes thus far have (unsurprisingly) revolved around the most apparent and obvious changes in our society that might account for this: more extreme beauty standards, perpetuated by social media, enforced upon young women to a greater degree than young men.

I would posit that this is an insufficient explanation for the phenomenon. Whilst I’m all for explanations of mental health conditions that invoke either social pressures or the vicissitudes of late capitalism, I’m uncertain that they tell the full story.

Moving targets: how classification changes conditions

In addition to the obvious leaps and bounds made in the destigmatisation of mental health conditions which mean that more people are willing to actively engage support networks, or that wounds are more likely to be identified (or identifiable) as self-inflicted, it might be wise to look to the work of Ian Hacking, who wrote an exceptionally good paper delivered to the British Academy in 2006, called “Kinds of People: Moving Targets”. In it, Hacking describes the ways in which humans alter their behaviour based on the ways in which that behaviour is classified and treated. He makes a case study of Dissociative Identity Disorder (previously known as Multiple Personality Disorder), where patients present with symptoms of having more than one personality. When it was first classified, the average patient would have two or three personalities; after a time, the number rose to seven or eight. Hacking argues that humans are “moving targets” for classification, and that mental disorders in particular tend to shift fairly rapidly. Essentially, whilst the underlying phenomenon might be the same in a patient who has what we might now think of as “depression” in 1900 vs today, the symptoms they present may well be extremely different.

This is, I think, a partial explanation of the rise in girls admitted to hospital with self-harm wounds. Self-harm, specifically the practice of cutting one’s body with razorblades of similar sharp implements, has become emblematic of depression over the last twenty years in a way that it wasn’t previously. 

Why is cutting specifically more prevalent?

It’s anecdotal, but I remember talking to my mother about depression a few years back, and whilst she had eventually come to accept nearly all of the other symptoms, she was still confused by the self-harming. She didn’t remember knowing anyone who had done it when she was young. Now, it might be that the stigma around mental health meant that she was unlikely to see anyone who had done it, and people were actually self-harming in similar numbers. But that explanation is belied by the evidence in front of us, which suggests that it’s a phenomenon in uptrend. I would instead suggest that self-harming of this particular kind simply was less prevalent in the 1960s or 1990s, and that the popularisation of the image of cutting as an expression of internal pain is responsible for the increased propensity of teens in particular to cut themselves.*

This kind of symptom choice can also be seen in the propensities of different demographics to suicide. Suicide rates are nominally lower amongst African American men and other ethnic minorities in the USA than they are amongst white men. One of the posited explanations for this is that social norms in African American culture are inimical to the idea of killing oneself, and so men who want to die are more likely to engage in reckless behaviours (drunk driving, extreme speeding) or put themselves in other situations where death is highly likely. Similarly, the suicide rate amongst women is again lower than it is in men, particularly in the USA. The reason for this isn’t that women don’t try to kill themselves - all the evidence suggests that if anything they have a higher number of “attempts” than men - but that the means by which they try to do so are less likely to result in a completed suicide. That is, men are more likely to shoot themselves (which is quite effective) whereas women are more likely to cut themselves or take an overdose (which are comparatively less effective). A skeptical reader might ask if men are not just more likely to own guns, to which the answer is yes, but that the effect is also seen in other countries where gun ownership is far lower. Instead, the most plausible explanation is that there are different social norms on men and women which affect the kinds of choices they make with respect to how they decide to try to kill themselves.

Social norms explain gender disparities in self-harm

I don’t believe it’s a stretch, then, to say that the reason that more girls are cutting themselves is because there is more of a norm amongst girls of cutting themselves. I don’t doubt that social media and beauty standards have something to do with it, but there’s also the stigma amongst men against engaging in any behaviours which outwardly express pain (unless that pain is expressed as rage). I would suggest that men who are depressed are far more likely to get into fights, or drink heavily, or punch a wall or otherwise damage property. The symptoms we express and the coping mechanisms we have are socially conditioned, not innate.

What do we do with this information? I don’t think the answer is to shame people for self-harming: it’s a mechanism of expressing pain or coping with depression, and I don’t necessarily think it’s intrinsically more harmful than any other means of doing so. I do, however, think that there ought to be more awareness raised of the risks of engaging in cutting, as well as safe practices to ensure you don’t end up in hospital. Most people self-harming do not want to die, and admission to hospital is usually an undesirable side effect rather than the aim. It would be worthwhile, for example, to inform young girls that if they are going to self-harm, then they ought to do so with something clean, and under no circumstances should they cut up their arms rather than across, because cutting vertically is far more likely to cause a wound that will bleed dangerously and be harder to close. They should also make sure that they have antiseptic of some kind (wipes will do), and plasters or similar for covering wounds.

Essentially, the net conclusion of this has to be that cutting is here to stay (at least for now), and the best thing we can do is to make it safer. Social media might be making the problem worse, but pinning the blame entirely on Instagram is lazy and doesn’t take account of the best social scientific research we have looking into what we might call “symptom contagion” and how mental health conditions shift through classification and changing social norms. 


*An important caveat: this explanation applies primarily to a specific kind of self-harm, i.e. cutting. I’m less confident of how this argument then impacts upon people’s propensity to engage in other kinds of self-harming behaviour. These behaviours, which have been documented over an extremely long period of time, are expressions of a wide variety of different emotions. They include“neurotic” types like biting nails, picking at the body, or extreme hair removal; or miscellaneous self-injury like repetitive head-banging (Menninger, 1935). I think there’s a plausible argument suggesting that self-harming behaviours manifesting in one way are unlikely to result in other different kinds of self-harm like head-banging, on the basis that people don’t associate them in their minds. I think, however, that practices like burning oneself with cigarettes, or other kinds of self-injury which involve direct instrumental contact with the skin, are likely to be on the rise too. I suggest this because of the apparent similarity between these kinds of behaviour, and I think the mechanism by which “symptom contagion” occurs is through close association between the cause (i.e. depression) and the symptom (self-harm). Self-injury through burning is, for some, quite closely associated with the idea of self-harm, and so is likely to increase concurrently. 

Further reading:

Hacking, Ian. "Kinds of people: Moving targets." Proceedings-British Academy. Vol. 151. OXFORD UNIVERSITY PRESS INC., 2007.

Menninger, K. (1935), "A psychoanalytic study of the significance of self-mutilation", Psychoanalytic Quarterly: 408–466

Going beyond "why is this true?": improving generic feedback on analysis

Episode 1 of PhDigital is here!

Episode 1 of PhDigital is here!