{
  "abstract": "Background Self-harm is common in young people and girls have consistently higher rates of self-harm than boys. This gender gap is widening and there has been a notable increase in the number of girls presenting with self-harm to both hospital and their GP. Understanding reasons for this gender disparity is important to inform both approaches to prevention and support services for young people. It is possible that there are gender differences in exposure to adversity or in reaction to adversity after exposure has occurred.This study used cross sectional data from the UK Millenium Cohort Study (MCS) to explore risk factors associated with self-harm in adolescents, SpecificallyExamined the extent to which boys and girls differed in level of exposure to risk factors.Examined whether boys and girls had different patterns and strengths of associations between risk factors and self-harm.Evaluate the extent to which differential risk factor exposure explained the difference in risk of self-harm between boys and girls.Methods We used data from the sixth wave of data collection from the MCS when the young people were around 14 years of age. We extracted data on gender as defined in the first wave of data collection, self-harm as a dichotomous variable from a single question in wave 6, sociodemographic data, and a range of psychosocial risk factors. All analytic models were weighted to take complex survey design and study attrition into account. A modified Poisson regression approach compared prevalence of risk factors and exposure-related relative risk between genders. Propensity score matching determined whether there was a residual gender difference in self-harm.Results We included 11,196 individuals with complete or imputed data (94% of all individuals in wave 6). The overall prevalence of self-harm was 15.4% with the relative risk of self-harm in girls compared to boys 2.68 (2.35–3.05). There were gender differences in risk factor patterns and in the strength of association between some risk factors and self-harm. When propensity score matching was used to remove the effect of differential exposure, the relative risk of self-harm in girls compared to boys was lower at 1.77 (1.47–2.12), leaving approximately two thirds of the gender difference not explained by the factors assessed.Discussion There are some key differences in risk factors between genders that could inform more gender sensitive services. Definitions used in the data, particularly of self-harm and a binary gender option, may have influenced the results.",
  "authors": [
    {
      "affiliations": [
        "School of Medicine, University of Leeds, Leeds, UK"
      ],
      "name": "Cathy Brennan"
    },
    {
      "affiliations": [
        "School of Medicine, University of Leeds, Leeds, UK"
      ],
      "name": "Emma Diggins"
    },
    {
      "affiliations": [
        "School of Medicine, University of Leeds, Leeds, UK"
      ],
      "name": "David Cottrell"
    }
  ],
  "title": "OP68 Using cohort data to explore gender differences in adolescent self-harm: lessons from the UK millennium cohort study",
  "uid": "eae3d0b0-220d-5c8e-9578-aa224319e78e"
}