Comorbid substance use in psychosis: Someone else’s problem?

Substance misuse is common among people who experience psychotic illness, with approximately one-third diagnosed with lifetime comorbid substance misuse (Toftdahl et al., 2016). Detecting and intervening in comorbid substance use is important because it is associated with poorer outcomes across a range of domains including increased rates of not only relapse and re-hospitalisation but also violence, suicide and all-cause mortality (reviewed in Darke et al., 2019). This significant negative prognostic indicator is frequently overlooked both in research, where people with comorbid substance use are often excluded from randomised controlled trials or other experimental studies, and in the clinical setting where substance issues may be considered to fall in the domain of drug and alcohol services. Furthermore, it is possible, even likely, that problematic use of many substances may go undetected in our severe mental illness populations because we do not see substance misuse as our core business, or because we are not exploring all possible drugs that may be in use. So, is it time for us to start to get better at detecting and intervening in substance use? For those clinicians who are motivated to enquire about comorbid substance use as part of a holistic assessment, what are the major substances that we should be inquiring about?

Likely, the first drugs that spring to mind in association with psychosis will be cannabis and methamphetamine. It is undoubtedly true that most cases of new-onset psychosis in the context of substance use in Australia, at least in those aged below 30, are related to cannabis, and/or to stimulants, predominantly methamphetamine. The potential for cannabis to increase risk for psychosis has been comprehensively researched and debated over recent decades. Meta-analytic evidence suggests that cannabis use is associated with increased risk for psychosis and in a dose-dependent manner (reviewed in Darke et al., 2019). High-potency cannabis (‘skunk’) contains high concentrations of Δ⁹-tetrahydrocannabinol (THC), the principal psychoactive constituent of cannabis. Use is associated with both a higher risk for psychosis and a younger age of its onset. Based on evidence that cannabis use is associated with an earlier onset of psychotic illness, it has also been argued that use may hasten the onset of psychosis in those who are otherwise vulnerable by dint of, for example, familial risk, genetic predisposition and/or environmental factors such as childhood trauma or social deprivation.

Methamphetamine is the most common illicitly used amphetamine, with an estimated 37 million users globally (Lappin and Sara, 2019). Use is predominantly of ‘ice’, the high potency crystalline form of methamphetamine. Methamphetamine and its active metabolite amphetamine have long been known to induce psychotic symptoms during use, likely due to their ability to powerfully increase synaptic dopamine. Prevalence of psychotic symptoms is high in both recreational and dependent users of meth/amphetamine and, just as is true for cannabis, there is evidence of a dose-dependent pattern. The extent to which methamphetamine and amphetamine use increases risk for development of psychotic illness is less clear. There is a well-documented subset of users in whom psychotic symptoms will not resolve following use, giving rise to brief (at least for some) methamphetamine-related psychosis (Lappin and Sara, 2019).

It is now well established that a high proportion of people with diagnoses of ‘drug-induced’ psychosis go on to receive a later diagnosis of schizophrenia. Our work has demonstrated that rates of transition to schizophrenia vary by substance: 34% for cannabis-induced psychoses and 22% for amphetamine-induced psychoses. Importantly, though, while the transition rate for alcohol-induced psychosis was much lower at 9%, alcohol was vastly more common overall due to the much higher number of people who use alcohol compared to other substances (Murrie et al., 2020). As a clinician then, you are several times more likely to encounter individuals who have developed psychosis related to alcohol use than related to either cannabis or methamphetamine use, particularly among those with an older age of psychosis onset.

Importantly though, high levels of comorbid psychosis and substance use are not explained only by substance use increasing the risk for development of psychosis. The reverse relationship also holds. In a large Danish cohort, there was found to be a highly increased risk for a new diagnosis of substance misuse in the years that followed the diagnosis of schizophrenia (Petersen et al., 2019). Importantly, while highest in the first 5 years post-schizophrenia diagnosis, the increased risk persisted even 10–15 years later. Polysubstance misuse was common at 16.9%, highlighting the need to enquire about the use of multiple substances. Explanatory models of this increased risk for substance misuse onset following schizophrenia include self-medication with substances to lessen symptoms of severe mental illness and exposure to a wealth of use-maintaining factors including impaired cognition, limited social networks, poor coping strategies and limited access to cessation support.

Just as clinicians should be aware of the need to assertively manage onset of psychotic symptoms experienced in the context of substance use, so too they should be vigilant for new onset of alcohol and/or substance misuse in the aftermath of a psychosis diagnosis in order to prevent decline into problematic use. So, what should an essential substance screen comprise?

An often-repeated maxim from medical training is common things are common. The most likely substance to be used by anyone with severe mental illness is nicotine with estimates of regular smokers ranging between 60% and 90%. Any substance use screen should include tobacco, not only because it has recently been indicated as a predictor of psychosis onset but also because of its significant contribution to premature mortality among severe mental illness populations and because smoking cessation interventions exist. Nicotine aside, substance use disorders among people with severe mental illness most often involve alcohol, followed by cannabis, then opioids and sedatives. A smaller, though still significant, number use psychostimulants (including methamphetamine, amphetamine, cocaine) and hallucinogens (LSD, psylocibin, PCP) (Toftdahl et al., 2014).

The high misuse of opioids and sedatives is worthy of note. While many of these drugs may be acquired illicitly, they are often – at least initially – prescribed for chronic pain or management of common symptoms of severe mental illness including anxiety, agitation and sleep disturbance. Opioids have the highest potential for dependence of all illicit drugs, with approximately one in four users of heroin becoming dependent. Other members of the opioid class include morphine, codeine, oxycodone, methadone and fentanyl. In recent years, the use and abuse of pharmaceutical opioids has become a major public health concern of epidemic proportion in North America and is increasingly problematic in Australia and Europe (Darke et al., 2019). Benzodiazepine prescribing in severe mental illness populations in Australia is not uncommon, despite guidelines warning of the harms of continuing use. Clinicians must be mindful of the dependence potential of these drugs in severe mental illness (SMI) populations who are vulnerable to the development of dependence for the reasons outlined above.

Finally, attention should be given to emerging drugs of concern such as synthetic cannabinoid receptor agonists. ‘Synthetic cannabis’ is the name given to a range of such drugs with highly variable active constituents. The term is, however, misleading as the drugs are pharmacologically distinct from THC and other cannabis-derived cannabinoids. Indeed, these drugs differ significantly in their effects, which are more in keeping with those observed in psychostimulant use (Darke et al., 2021). Importantly, many urine or blood toxicology screens will not identify synthetic cannabis as cannabis and so taking a detailed clinical history is essential.

Individuals with comorbid substance misuse and psychosis present a number of treatment challenges, then, and a comprehensive assessment will extend far beyond use of cannabis and methamphetamine. Once aware of the extent of any substance use and related problems, important next steps include understanding the relationship between substance use and the experience of symptoms, and the individual’s motivation to seek treatment for either issue. Effective interventions for those who are motivated do exist: group counselling, contingency management and residential treatment. Realistically, the path to recovery may not be smooth and relapse will be likely. Barriers to effective care include poorer adherence to treatment, poorer engagement in services, and a pattern of being difficult to reach, then presenting in crisis. Treatment of both conditions concurrently is the gold standard, but the availability of integrated treatment interventions in mainstream public health services is unfortunately limited. Nonetheless, if we as clinicians are sufficiently motivated to detect and discuss comorbid substance use routinely, we will be better serving at least one in three of those in our care.