Early experiences of physical and sexual abuse are considered to contribute to the development of psychosis later in life (Read et al. Reference Read, van Os, Morrison and Ross2005; Schreier et al. Reference Schreier, Wolke, Thomas, Horwood, Hollis, Gunnell, Lewis, Thompson, Zammit, Duffy, Salvi and Harrison2009; Arseneault et al. Reference Arseneault, Cannon, Fisher, Polanczyk, Moffitt and Caspi2011). Similarly, cannabis use has been linked to an increased risk of psychosis and an earlier age of onset of psychotic illness (Moore et al. Reference Moore, Zammit, Lingford-Hughes, Barnes, Jones, Burke and Lewis2007; Large et al. Reference Large, Sharma, Compton, Slade and Nielssen2011). A growing number of studies have shown that childhood trauma and cannabis use can combine more-than-additively to produce a highly elevated risk of psychotic symptoms and psychosis (e.g. Houston et al. Reference Houston, Murphy, Adamson, Stringer and Shevlin2008; Harley et al. Reference Harley, Kelleher, Clarke, Lynch, Arseneault, Connor, Fitzpatrick and Cannon2010).

Most recently, two studies published in Psychological Medicine have shown that: (i) psychotic symptoms are particularly likely to occur amongst cannabis users with a history of childhood physical or sexual mistreatment (Konings et al. Reference Konings, Stefanis, Kuepper, de Graaf, ten Have, van Os, Bakoula and Henquet2011), and (ii) that psychosis risk is enhanced amongst cannabis users who have experienced non-consensual sex before the age of 16 years (Houston et al. Reference Houston, Murphy, Shevlin and Adamson2011). Crucially, neither study considered the possibility that the risk posed by prior trauma could be enhanced amongst cannabis users because individuals in this group are also likely to have taken illicit psychostimulants implicated in the development of psychosis (Fergusson & Horwood, Reference Fergusson and Horwood2000; Lynskey et al. Reference Lynskey, Heath, Bucholz, Slutske, Madden, Nelson, Statham and Martin2003; Curran et al. Reference Curran, Byrappa and McBride2004; Barnett et al. Reference Barnett, Werners, Secher, Hill, Brazil, Masson, Pernet, Kirkbride, Murray, Bullmore and Jones2007). To test this idea, I examined data from 7125 participants drawn from the Adult Psychiatric Morbidity Survey 2007, as utilized in Houston et al. (Reference Houston, Murphy, Shevlin and Adamson2011) .

As shown in Houston et al. (Reference Houston, Murphy, Shevlin and Adamson2011), cannabis use moderated the link between non-consensual sex in childhood and psychotic disorder in the last week [odds ratio (OR) 10.53, 95% confidence interval (CI) 1.14–99.64; all analyses were adjusted for age, gender, ethnicity, education, employment, alcohol use, sexual trauma after age 16 years, presence of neurotic disorder, and, where appropriate, cannabis and stimulant use]. Similarly, stimulant use (cocaine, ecstasy or amphetamines) combined with non-consensual sex before the age of 16 years to predict a raised risk of psychosis (OR 16.75, 95% CI 1.79–157.2). Cannabis dependency levels did not interact with non-consensual sex to predict psychosis, suggesting that psychostimulant use is not merely a proxy for the effects of high levels of cannabis consumption.

Critically, adjusting for the under-16 sex×stimulant use interaction removed the link between the under-16 sex×cannabis use interaction and psychosis (OR 8.22, 95% CI 0.48–140.9). As anticipated, the link between under-16 sex and psychosis amongst cannabis users (OR 17.43, 95% CI 2.61–116.43; illustrated in Fig. 1 a) was found to be attributable to a strong link between non-consensual sex in childhood and psychosis amongst the 37.1% of cannabis users who have also taken stimulants (OR 77.7, 95% CI 10.39–581.1 in age and gender adjusted analysis; OR 70.71, 95% CI 1.38–3631 in fully adjusted analysis), as shown in Fig. 1 b. Early non-consensual sex was unrelated to psychosis amongst those who had used cannabis alone.

Fig. 1. Relationship between non-consensual sex before the age of 16 years and psychosis for: (a) those who have used cannabis and those who have not taken cannabis or stimulants (cocaine, amphetamines or ecstasy); and (b) cannabis users who have taken stimulants, cannabis-only users and those who have not taken cannabis or stimulants. Note: participants who have used stimulants but not cannabis (approximately 1% of the sample) were excluded for illustration purposes.

Taken together, these findings suggest that stimulant use amongst cannabis users rather than cannabis use alone may enhance the impact of childhood trauma on the likelihood of developing psychotic disorder. Future studies examining cannabis×trauma interactions in psychosis should include the main effect of stimulants and interactions between trauma and stimulants in their analyses. Such an approach may uncover novel relations and will ensure that those examining synergistic relations between cannabis use and adverse experiences accurately identify the specific drugs that contribute to psychosis risk.