Unravelling the paradox of clozapine and mortality in treatment-refractory schizophrenia using linked electronic health records

Schizophrenia is chronic mental disorder characterised by continuous or relapsing episodes of psychosis. During these episodes individuals experience include hallucinations (typically hearing voices), delusions and disorganised thinking. It can be a debilitating condition for patients and their families, with significant wider costs to society. Individuals with schizophrenia have an estimated 2.5-fold increased risk of premature death compared to the general population and die 15-25 years earlier, mainly due to cardiovascular disease and suicide. This situation is even worse for approximately one quarter of patients with schizophrenia whose illness does not get better with standard antipsychotic medicines: so-called treatment refractory schizophrenia (TRS). Clozapine is the only antipsychotic that is known to be effective in TRS, and is recommended by the National Institute for Health and Care Excellence (NICE) for all TRS patients.

Given that clozapine is usually prescribed in patients with more severe illness, and clozapine is known to cause a range of potentially fatal side effects, it might be assumed that treatment with clozapine would increase the risk of premature death. Remarkably, the opposite appears to be the case. Several large studies have reported that people who take clozapine are substantially LESS likely to suffer premature death than individuals on other antipsychotics. Understanding this apparent paradox is crucial, and is the focus of this fellowship. Is clozapine so much more effective at treating schizophrenia that this outweighs the increased risk of side effects? If so, how is clozapine helping? Is it by reducing suicide, as some studies have suggested, or are other mechanisms at play? Alternatively, were the previous studies which showed this effect biased or otherwise flawed? Will our studies, which will avoid the pitfalls of earlier research, reach the same conclusions?

Using three electronic health records (EHRs), I will investigate the relationship between clozapine and premature mortality risk to see if previous findings showing that clozapine prescription in TRS patients provides a protective effect on mortality, can be replicated. I will design these studies in such a way to avoid the weaknesses of previous studies. I will also investigate causes of death in patients treated with clozapine, and quantify the protective and detrimental effect of clozapine for these causes of death, to establish if the long-term mortality benefits of clozapine outweigh associated physical health risks. I will then explore the ways in which clozapine might exert its mortality reducing effects. The EHRs selected span the full care spectrum to offer breadth and richness of information. They will be linked to national clozapine registries, which oversee the prescribing of clozapine in secondary care.

If a clozapine is confirmed to provide a mortality advantage, then wider and earlier use may reduce the mortality gap and health inequalities that exist. Findings could provide reassurance that clozapine's physical health side effects are outweighed by its mortality benefits, and that it should be considered at the earliest opportunity for patients who meet TRS criteria. Our results may also identify some groups for whom the benefits do not outweigh the risks, and this is important information. Findings may support a case for a review of clozapine treatment guidelines, for restrictions on use to be re-evaluated, and for clozapine to be recommended in some patients but not in others. Elucidating mortality rates from specific causes, would enable prioritisation of specific prevention and intervention strategies.

Aim
To investigate if clozapine provides a protective effect on mortality in treatment refractory schizophrenia (TRS) in the UK, addressing the methodological weaknesses identified by previous studies. I aim to examine moderators and mediators of this effect, as well establish the main causes of death in clozapine-treated patients.

Study 1: Clozapine prescription and premature mortality risk

Objectives
1.1 To test the hypothesis that, compared to other antipsychotics, clozapine prescription (at time of death) provides a protective effect on mortality
1.2 To address the methodological shortcomings of previous studies, avoiding as far as possible the biases/confounding listed in Table 1 which have affected previous estimates
1.3 To investigate the impact of other aspects of clozapine prescribing i.e dosage/plasma concentration/ cumulative exposure, on premature mortality risk
1.4 To investigate mortality immediately following clozapine cessation

Study 2: Cause-specific mortality and clozapine prescription

Objectives
2.1 To compare causes of death in patients prescribed clozapine versus other antipsychotics
2.2 To quantify protective and detrimental effect of clozapine for causes of death identified in 2.1 to establish if the long-term mortality benefits of clozapine outweigh associated physical health risks
2.3 To study risk factors for specific causes of death identified


Study 3: Exploration of moderators and mediators of the mortality reducing effects of clozapine

Objectives
3.1 To determine the extent to which any protective effect of clozapine on premature mortality is mediated or moderated by:
i. Better adherence to treatment for cardio-metabolic risk factors (BP/HbA1c/cholesterol/BMI)
ii. Modulatory role for effectiveness/safety (harms/benefits) of co-prescription of medications for multimorbidity
iii. Adherence to preventative strategies (i.e. cancer screening/Covid vaccination/lifestyle behaviours including smoking/alcohol use)
iv. Non-fatal suicide behaviours
3.2 To investigate the relationship between (i) to (v) above, and specific causes of death.