We appreciate Liu's interest in our case–control study of DSM-IV Axis II personality disorders and suicidal behaviour in China; however, the commentary does not accurately represent the purpose of the report or our conclusions.

Liu was critical of our focus on personality disorders, stating that ‘the authors do not adequately explore possible environmental reasons for their findings […] DSM-IV personality categories may not be appropriate tools for understanding suicidal behaviour in China’. One possible explanation for the much lower reported prevalence of mental disorders in suicide and attempted suicide in China compared with high-income countries ^{Reference Yang, Phillips, Zhou, Wang, Zhang and Xu1,Reference Jiang, Li, Phillips and Xu2} is the failure to consider Axis II personality disorders in the relevant studies. One of the goals of this study was to test this hypothesis by examining the relationship between personality disorders and suicidal behaviour in China using case–control methodology. We did not make an a priori assumption about the role of such disorders in suicidal behaviour. In fact, as stated in the introduction of the paper, ‘The purpose of the current study is to address the gaps in data on personality disorders and suicidal behaviour in China’.

We found that a low percentage of suicide decedents (7%) and suicide attempters (6%) in China had one or more Axis II personality disorders, and that a very low percentage of living community controls (1%) had such a disorder. Despite some concerns about the sensitivity of available instruments for measuring personality in China and about respondents' reluctance to report negative traits, our conclusion was that ‘personality disorders as defined by DSM-IV are much less prevalent in mainland China’. This result confirms our initial impression that failure to consider the categorically defined DSM-IV personality disorders is not the main reason for the previously reported low prevalence of mental disorders among individuals with suicidal behaviour in China. ^{Reference Phillips3} Of course, it remains possible that using dimensional measures of personality traits (rather than the categorical classification of personality disorders) could identify personality characteristics that are important predictors of suicidal behaviour in China. Prior research on the relationship of dimensional measures of impulsivity to suicidal behaviour in China ^{Reference Zhang, Xiao and Zhou4,Reference Ai, Chen, Wang, Gan, Shen and Li5} is an example of the type of work needed to assess this possibility.

Liu suggested other targets for research and prevention that should take precedence over the examination of Axis II personality disorders, including environmental influences, culture and economic factors. We agree that these are potentially important foci, but they were intentionally not considered in the current report that focused on the relative importance of Axis I and Axis II mental disorders as predictors of fatal and non-fatal suicidal behaviour. These other factors will be addressed in subsequent reports on this study.