Article Text
Statistics from Altmetric.com
Question
Question: What is the risk of substance-induced psychosis (SIP) converting into schizophrenia spectrum disorder?
Population: All 18 478 individuals (81% men, mean age 43.7 years) in Finland with a first hospital admission for SIP between 1987 and 2003. Individuals were identified using the Finnish Hospital Discharge Register, and were followed until first occurrence of a schizophrenia spectrum disorder, death or the end of December 2003, whichever occurred first.
Setting: Finland; January 1987–December 2003.
Prognostic factors: A discharge diagnosis of SIP between 1987 and 2003 (based on DSM-III-R codes prior to 1995 and ICD-10 codes thereafter). The substance associated with the SIP admission was recorded where available and adjusted for in analyses. Additional factors considered as covariates were age, sex and duration of initial hospitalisation for SIP.
Outcomes: Conversion to a discharge diagnosis of a schizophrenia spectrum disorder during follow-up to December 2003 (DSM-III-R codes for diagnoses prior to 1995 and ICD-10 codes thereafter). Conversion rate was calculated by dividing the number of cases by the corresponding follow-up time, and cumulative probabilities for conversion were calculated using the Kaplan-Meier method.
Methods
Design: Prospective cohort study.
Follow-up period: Eight years following first admission for SIP.
Main results
Of the 18 478 patients discharged after a first hospital admission for SIP, the majority (15 787, 85.4%) had a diagnosis of alcohol-induced psychosis (83% men, average age 45.1 years); 825 (4.5%) had amphetamine-induced psychosis (74% men, average age 26.1 years); and 215 (0.7%) had cannabis-induced psychosis (88% men, average age 23.4 years). Patients with alcohol-induced psychosis were significantly older than patients with either cannabis-induced psychosis or amphetamine-induced psychosis. Individuals diagnosed with cannabis-induced psychosis had the highest risk of converting to a schizophrenia spectrum disorder during follow-up (conversion probability 46%, 95% CI 35% to 57%). This represents a crude conversion rate of 12.5/100 person-years (95% CI 8.9 to 16.1). Amphetamine-induced psychosis was associated with a 30% conversion probability (95% CI 14% to 46%), and a crude conversion rate of 4.5/100 person-years (95% CI 3.8 to 5.3). Alcohol-induced psychosis was associated with a 5% conversion probability (95% CI 4.6% to 5.5%), and a crude conversion rate of 0.65/100 person-years (95% CI 0.6 to 0.7). Older age at SIP diagnosis was significantly associated with conversion to schizophrenia, with patients over the age of 30 years significantly less likely to be diagnosed with schizophrenia during follow-up than patients under 30 years at SIP diagnosis (HR 0.38, 95% CI 0.33 to 0.45, p<0.0001). Patients with an index hospitalisation for SIP lasting longer than 7 days were also significantly more likely to have a diagnosed schizophrenia spectrum disorder during follow-up. Gender was not significantly associated with conversion risk.
Conclusions
The likelihood of substance-induced psychosis converting to a schizophrenia spectrum disorder varies depending on the substance that precipitated psychosis hospitalisation.
Notes
Results for substance-specific risks adjusted for patient age were not reported.
Commentary
A PubMed search for ‘epidemiology and schizophrenia’ currently yields over 11 400 citations, compared with just 417 for a similar search of substance-induced psychoses (SIPs, typically defined as acute psychotic episodes following psychoactive substance use, with symptoms quickly remitting following elimination). This difference in research investment remains incongruous, even when the relative difference in incidence is considered; schizophrenia incidence is around eight times greater than SIPs.1
The recent study by Niemi-Pynttari and colleagues illuminating the epidemiology of SIPs, is the largest of its kind to date. Using Finnish national register data, the authors identified over 18 000 people with a first admission for SIPs between 1987 and 2003. Participants were traced over 8 years to estimate the transition to schizophrenia spectrum disorders (SSD), where possible, by primary substance involved. Somewhat surprisingly, alcohol-induced psychoses accounted for 85% of all SIPs. By comparison, just 0.5% of SIPs were primarily attributed to cannabis. However, patterns of SSD transition were reversed, with a cumulative probability of transition to SSD of 46% for cannabis-induced psychoses compared to just 5% in alcohol-induced psychoses.
The authors’ work has important public health implications. By applying the crude incidence of schizophrenia in Finland for this period2 to their data, 96.7% of SSDs in the SIPs group could be obviated if substance-induced psychoses could be prevented, assuming causation (Supplemental table 1). At the population level, this corresponds to 5.4% of all schizophrenia cases in Finland. Notably, these figures varied by substance, with the greatest population-attributable risk associated with alcohol-induced psychoses (3.1%). Among those with an SIP, however, only eight cannabis-induced psychoses would need to accrue for one person to develop an SSD which would not have otherwise occurred, compared with 162 for alcohol-induced psychoses. This work represents a significant advance for the field and can be used to highlight the public health impact of substance-induced psychoses.
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Files in this Data Supplement:
- Data supplement 1 - Online references
- Data supplement 2 - Online table
Footnotes
-
Sources of funding: The Finnish Psychiatric Association, the Finnish Society of Addiction Medicine and the Academy of Finland.
Footnotes
-
Competing interests None.