BACKGROUND: Several epidemiological studies have shown that cannabis; the most
widely used illegal drug in the world, is associated with schizophrenia spectrum 
disorders (SSD).
AIMS: To assess the characteristics of cannabis use and its association with SSD 
in a cohort of psychiatrically ill patients and discuss the implications for
policy development
METHODS: This is a retrospective analytical study of a cohort of psychiatric
patients who received treatment in the psychiatry unit of the Provincial General 
Hospital, Ratnapura, Sri Lanka over five years (2000 - 2004). The schizophrenia
spectrum disorders defined in this article include schizophrenia and the
schizoaffective disorders.
RESULTS: A total of 3644 patient records were analyzed. The percentage of self
reported life time cannabis (LTC) use was 2.83% (103, all males). Sixteen percent
(576) of the total cohort was diagnosed with SSD by 2009. Male sex and LTC use
were significantly associated with SSD (p < 0.01 and 0.001 respectively). In the 
majority (91.5%), cannabis use preceded the diagnosis. There were 17(16.5%)
patients diagnosed as cannabis induced psychosis and 7 (41.2%) of them were
subsequently diagnosed as SSD. This group was significantly more likely to have
had a past psychiatric consultation, but other demographic and clinical
correlates did not differ from the rest of the LTC users.
CONCLUSIONS: Self reported LTC use was strongly associated with being diagnosed
with SSD. However we could not identify a particular subgroup of users that are
at increased risk to recommend targeted primary prophylaxis. The policy
implications of this observation are discussed.