We read with interest the recent study by Henquet and colleagues. ^{Reference Henquet, van Os, Kuepper, Delespaul, Smits and Campo1} As well as providing further support for the well-established theory that cannabis may worsen or re-awaken psychosis in vulnerable adults, this study reports the fascinating and novel finding that cannabis appears to differentially affect mood – with patients with a psychotic disorder, but not controls, reporting improvements in negative affect following cannabis use. On the other hand, cannabis enhanced positive affect in patients and controls alike.

Previous studies have been contradictory regarding the effects of regular cannabis use on mood. Denson & Earleywine found that regular users reported less depressed mood and more positive affect than non-users, ^{Reference Degenhardt, Hall and Lynskey2} whereas Degenhardt and colleagues reported that heavy cannabis use and depression were associated. ^{Reference Denson and Earleywine3} The reason for these differences is not clear, but may be due to differences in cannabis composition, as pure delta-9-tetrahydrocannabinol is anxiogenic when given acutely, whereas cannabidiol appears to ameliorate these effects. ^{Reference Bhattacharyya, Morrison, Fusar-Poli, Martin-Santos, Borgwardt and Winton-Brown4}

The finding that patients derived more benefit from cannabis use in terms of mood suggests that the association of early cannabis use with subsequent onset of psychosis may not, in fact, be a causative relationship as previously reported. ^{Reference Moore, Zammit, Lingford-Hughes, Barnes, Jones and Burke5} Rather, early cannabis use in these (already vulnerable) individuals may be more likely as they derive more benefit – in terms of mood enhancement – than individuals who are not at risk of psychosis. Henquet and colleagues also report that the effects on mood are acute, whereas effects on psychosis are subacute. It would be interesting to determine whether the effects on mood and psychosis occur with equal frequency earlier in the illness, because if psychosis emerges only with repeated dosing, this may be a further maintaining factor in early use.

Regardless of the aetiological relationship of cannabis use to psychosis onset, this study highlights an important point – people take cannabis because they feel that they derive benefit from it, and patients with psychosis are no different in this respect. In terms of clinical practice, this paper highlights one reason why service users may continue to smoke cannabis, despite the fact that it clearly worsens their psychotic symptoms. This awareness can add to our understanding and attitude towards the service user, and enable us more creatively to help the service user find alternative ways to boost their mood.