To the Editor

We welcome Richard Liu's analysis of the 2007 Adult Psychiatric Morbidity Survey (APMS), which provides important insights, but question his claim to present the first epidemiological study of non-suicidal self-injury (termed self-harm or non-suicidal self-harm in the UK) (Liu, Reference Liu2021). The survey is, in fact, the second in a series carried out in England in 2000 (Singleton, Bumpstead, O'Brien, Lee, & Meltzer, Reference Singleton, Bumpstead, O'Brien, Lee and Meltzer2001), 2007 (McManus, Meltzer, Brugha, Bebbington, & Jenkins, Reference McManus, Meltzer, Brugha, Bebbington and Jenkins2009), and 2014 (McManus et al., Reference McManus, Hassiotis, Jenkins, Dennis, Aznar, Appleby and Brugha2016), recently summarised in a Lancet Psychiatry paper (McManus et al., Reference McManus, Gunnell, Cooper, Bebbington, Howard, Brugha and Appleby2019). In neither the title, the abstract, nor the introduction does Liu mention that the data he used were collected in 2007, 14 years ago. He therefore misses important changes in the epidemiology of self-harm that have occurred since then, including shifts in the prevalence, methods, and characteristics of those who engage in self-harm.

Our primary analyses of the same dataset, alongside the surveys in the APMS series carried out earlier (2000) and more recently (2014), reveal that the prevalence of non-suicidal self-harm in England increased steeply from 3.8% in 2007 to 6.4% in 2014. While the rise was evident across age groups and in both sexes, it was particularly pronounced in young women, in whom the rate rose from 11.4% in 2007 to 19.7% in 2014. The comparable rate in young men in 2014 was 7.9%. Liu's discussion of the lack of difference in rates by sex does not reflect the current situation: non-suicidal self-harm is now more common in women than in men.

Since 2007 both the methods of self-harm and reported motivations have also changed. Cutting has come to account for a higher proportion of female cases, and people have become more likely to report self-harm as being a coping mechanism. In addition, the more recent 2014 survey included improved methods of enquiry, establishing a history of self-harm in the past year as well as over the lifetime, enabling a more accurate estimate of the proportion who were not in contact with services.

Liu's article does deal with several other issues of interest and value. However, we would welcome a correction from the author that alerts readers to the distinct and well-documented changes seen in the epidemiology of self-harm in the general population since these data were collected.