Substance Misuse and Early Psychosis

PREVALENCE

The rate of substance abuse among populations experiencing their first episode of psychosis is higher than expected. The Age-Beginning-Course (ABC) Study in Germany found a twofold increase in rate of substance abuse in first-onset schizophrenia (alcohol abuse rate 24%, other drug abuse 14%), compared with the general population. 1 In a British study, the rate of alcohol/drug abuse plus drug use in first-presentation psychosis was 37%. 2 Verdoux et al. in France found 33% of first-admission persons with psychosis to have substance abuse. 3 In Canada, the rate was found to be 44% in 2004 4 and the same group found a rate of 51% in another sample in 2007. 5 In Melbourne, the rate for Early Psychosis Prevention and Intervention Centre (EPPIC) entrants was 62%. 6 It appears that these rates may be gradually increasing over time. Low rates (9%) in a Singapore study were ascribed to local factors, including strict drug laws. 7 Prodromal states in individuals also have high rates of substance abuse. 8

Use of more than one substance is the rule, and prominent among these in international samples is cannabis and alcohol. This is the case in Australian and New Zealand samples 6 as well. There is also a gender effect in that males are more likely to abuse substances, which is of course the situation in the general population.

THE RELATIONSHIP BETWEEN SUBSTANCE USE AND PSYCHIATRIC SYMPTOMS

Associations between substance abuse and illness related factors are well known in the dual diagnosis literature, generally being in the direction of worsened course and prognosis. In people with first-onset psychosis, substance abuse was associated with increased risk of readmission and presenting with psychotic symptoms over a 1–2 year follow-up period. 9 , 10 Illness features at onset do not differ between those who use cannabis and those who do not. 11 Cannabis use was associated with more positive symptoms and a more continuous illness over a 4-year follow-up period in recent onset psychosis. 12 An association of particular relevance in the present context is the earlier onset of psychosis among those patients with a lifetime history of substance misuse. 4 , 5 , 13 A British group found that a lifetime history of substance abuse at first presentation for psychosis was associated with a 2–3 years earlier onset of schizophrenia, although no increase in duration of untreated psychosis. 14

Rates of parasuicide are increased in early psychosis when substance abuse is present. This is thought to be a reflection of the association between substance abuse and suicide seen in the general population. 3

There have been varying results with regard to associations with other specific features of psychosis. In one study, cannabis was associated with more positive symptoms and depression. 5 A British group was unable to identify any appreciable differences in symptom pattern between patients with schizophrenia who were and were not cannabis users. 11

COGNITIVE IMPAIRMENT

As with populations without schizophrenia, there is variability in whether cognitive impairment ensues following substance abuse, this effect being seen more when the abused substance is cocaine rather than alcohol or cannabis. These issues are reviewed by Pencer and Addington. 15 Over a follow-up period of 2 years, they found that substance abuse did not bear any relationship to cognitive performance in those with first-onset psychosis. These results were confirmed in a study by Barnes et al. 16 A group in New York, on the other hand, found that in first-onset dual diagnosis, patients had higher parental social class and better cognitive and language skills. 16 It has also been found that use of cannabis before illness onset is associated with marked ‘sparing’ of neurocognitive function at the 10-year follow up. 17

CAUSAL RELATIONSHIPS

Analysis of the causal relationship of substance abuse and early psychosis begins with an examination of the temporal relationships of the two disorders. Clearly for any factor to be a cause it must precede whatever it is thought to be causing.

Hambrecht and Hafner 1 were able to divide their group of first-onset drug (not alcohol) abusing psychotics into three groups of similar size: 28% in which drug abuse began 1–5 years before the first sign of schizophrenia, 36% in which they began more or less simultaneously, and 38% in which it began afterwards. As the drug abused was predominantly cannabis, this led the authors to postulate three main mechanisms of interaction for cannabis and mental illness: ‘vulnerability’ to schizophrenia in which cannabis causes a chronic deterioration of mental health increasing the likelihood of developing psychosis; ‘stress’ in which cannabis acts as a dopaminergic stressor precipitating psychosis; and cannabis as a ‘coping’ strategy to deal with symptoms of psychosis (or medication side effects). 18 The last is also sometimes spoken of as ‘self-medication’ although this term is misleading as drugs of abuse are generally inappropriate as medication and mostly do not have effective treatment outcomes in schizophrenia.

CANNABIS

Psychotic symptoms are twice as common (even correcting for confounding influences) in young people who use cannabis. 19 Cannabis use has been found to be more likely to produce concurrent psychotic symptoms in ‘psychosis prone’ young people, who are also more likely to report perceived hostility. 20 There is a dose–response effect between cannabis use and psychotic symptoms in high-risk and normal subjects. 21 However, in a high-risk group for schizophrenia, cannabis use was not found to have a significant impact on precipitating psychotic illness over a 12-month follow-up period. 22

The considerable body of evidence for cannabis as a part-cause of schizophrenia in the longer term was reviewed by Arseneault et al. 23 They included in their review the frequently cited prospective studies of Swedish conscripts who developed schizophrenia in proportion to the amount of cannabis they used; 24 the Netherlands Mental Health Survey and Incidence Study (NEMESIS) in which the development of psychotic symptoms over 3 years was related to baseline cannabis use; 25 and the Dunedin birth cohort study in which later development of psychotic symptoms and schizophrenia was associated with adolescent cannabis use. 26 Arseneault et al. concluded that cannabis use played some causal role and doubled the risk of developing schizophrenia in the long term. The risk appears to be in those already vulnerable by virtue of pre-existing psychotic symptoms. The population-attributable fraction of schizophrenia due to cannabis is at least 8%. Neurobiological explanations for the association include the differential risk of developing schizophrenia after cannabis associated with catechol-O-methyl transferase gene polymorphisms. 27

METHAMPHETAMINE

The prevalence of psychosis (as assessed by a psychosis screener) in young methamphetamine users in Australia is high (13%) and is mostly diagnosed as schizophrenia or drug-induced psychosis. 28 As the prevalence of amphetamine use among the general population and school students is 2–3%, 29 the relevance of this drug to first-onset psychosis may be substantial despite its lesser prominence in first-onset samples. While it is widely accepted that amphetamines are a sufficient cause to produce psychosis, which is usually transient, 30 this is by no means a universal outcome for methamphetamine users and factors have been identified which predispose to the development of psychosis. These include younger age at commencement of drug use, heavier use, presence of schizoid and schizotypal traits, depression, alcohol dependence and antisocial personality disorder. 31 Methamphetamine users who develop psychosis, especially when this is prolonged, are considerably more likely to have family members with schizophrenia, 32 suggesting that the ‘vulnerability’ model may be relevant for this drug.

ALCOHOL

Relatively little attention has been given to alcohol despite its high level of use among early onset psychosis dual diagnosis cases. It appears to be considerably less associated with relapse than other drugs. 10

DRUG-INDUCED PSYCHOSIS

The concept of acute ‘drug-induced psychosis’ should be regarded with caution. When schizophrenia and substance abuse co-exist, it has been found that in about half of the cases the original diagnosis was of a drug-induced psychosis, which may have delayed treatment for schizophrenia. 33 This diagnosis was made in 9% of a first-onset psychosis sample but diagnoses were in 30% of cases reassigned over a 3-year period. 2 In a study of patients in Scandinavia who were first diagnosed with cannabis-induced psychosis (which had a very low incidence), further psychotic episodes over ensuing years were usual and almost half the cases were subsequently diagnosed with a schizophrenic spectrum disorder. 34

INTERVENTIONS

Specific interventions for dually diagnosed first-onset psychosis cases typically include components of motivational enhancement, psychoeducation, skills training and support, taking into account stage of recovery. In a program in Calgary, Canada, in which dual diagnosis was addressed as part of a broader early psychosis program, there was a large drop out rate of around 50%. However, among those that remained, the level of substance abuse decreased significantly over 1 year. Specific substance abuse groups were unpopular in the early stages of engagement so that the interventions were integrated into the broader program. 35 These outcomes were confirmed in a later study by the same investigators with larger subject numbers followed up over 3 years. 5 In a Queensland study, a substance abuse treatment component integrated into a more general program delivered to first-onset cases identified at hospital admission 36 resulted in better substance use and readmission outcomes than did the unenhanced program over 1 year; 37 a similar program in Sydney resulted in decreased substance use and improved self-efficacy. 38

The rate of substance abuse for participants in the overall EPPIC program halved over 18 months; those who reduced use had better outcomes than those who had never abused substances. 6

Over a longer period of time after induction into the EPPIC program, of those subjects who could be located and recruited an average of 8 years later, 50% still had active psychotic disorders and 23% of these had an additional current substance use disorder, which was less than half their lifetime substance use disorder prevalence. This suggests some eventual remission of the disorder. 39 The effect of program intervention was unclear. The EPPIC first-episode program in Melbourne now includes a cannabis-specific component. 40

CONCLUSION

There is a high prevalence of substance misuse among persons with early psychosis, with cannabis and alcohol featuring prominently. Substance misuse is associated with earlier onset and possibly more positive symptoms, although apparently not with greater cognitive impairment. Cannabis appears to confer increased likelihood of developing schizophrenia in biologically vulnerable individuals. Amphetamines also cause psychosis which may become chronic, although specific vulnerability to this effect is less well established. Many cases of so-called ‘drug-induced psychosis’ become diagnosed as schizophrenia in later years. Specific intervention programs report positive outcomes with regard to substance misuse and course of psychosis.