Background Male suicide rates increased remarkably in Scotland in the 1980s and 1990s, with higher rates in more deprived areas. We examine trends in male suicide in Scotland from 1980 to 2015 by method of suicide and individual socioeconomic position. We also explore trends in inequalities by individual socioeconomic position and the extent to which this may be attributable to suicide method.

Methods Suicide deaths for 1980–2015 were obtained from National Records Scotland. Inequalities were assessed using National Statistics socio-economic classification (NS-SEC) for ages 20–59 in 2000–2002 and 2010–2012. Standardised death rates were calculated by age with European population 2013 as reference, and Poisson regression was used to assess trend significance. Inequalities were assessed using Slope of Index of Inequality (SII).

Results There were 12 281 suicide deaths between 1980 and 2015. No significant linear trend was observed over time for all ages, but suicide rates per 1 00 000 person-years increased from 21 to 27 between 1980 and 2002 (p<0.001), and decreased from 27 to 20 from 2002 to 2015 (p<0.001). No significant trend was observed for poisoning, but suicide rate by hanging, suffocation and strangulation increased over time (p<0. 001), and decreased for other suicide (p<0.001). There were significant differences in suicide rates between the NS-SEC groups, and by suicide method (p-values<0.001). SII: 85 (95% CI, 77 to 92) in 2000–01 and 62 (95% CI, 55 to 68) in 2010–12 per 100 000 person-years, indicated that inequalities between social class extremes were significantly higher in 2000–02 than 2010–12. Suicide by hanging, suffocation and strangulation accounted for 44% of inequalitiesin 2000–02 (SII=37 (95% CI, 33 to 42)) and 49% in 2010–2012 (SII=30 (95% CI, 25 to 34)).

Conclusion The decline in male suicide rates may be attributed to suicide prevention strategies introduced by the Scottish Government from 2002 such as Choose Life. Despite decreasing trends of male suicide, suicide by hanging, suffocation, and strangulation increased over time. Inequalities by individual deprivation decreased between 2000–02 and 2010–12. Limitations are that NS-SEC categories, never worked and long term unemployed and not classified, were excluded from the analysis assessing inequalities since the interpretation of results for this group was ambiguous. What effect this omission would have on the estimation of inequalities is unclear. Caution should be taken when comparing inequalities by NS-SEC 2001 and 2011 as in 2011 NS-SEC category was estimated for those without occupation. Policy should be directed at reducing deaths from hanging, suffocation, and strangulation.