Clozapine is a double-edged sword. On one side, it’s a powerful antipsychotic that can help people with treatment-resistant schizophrenia manage their symptoms (Schneider-Thoma et al.,2025; Murray et al, 2025). On the other hand, it is notorious for side effects, from heart inflammation to severe blood disorders, that make doctors break into a cold sweat (Espejo & Kohrassani 2025). And now, new research from Hong Kong adds another twist: clozapine may also increase the risk of infections, especially in older patients (Hu et al., 2025).
This isn’t just a quirky medical fact; it’s a reminder of the tightrope clinicians walk when prescribing potent medications. Because let’s be honest, trading psychiatric stability for a higher chance of catching the flu isn’t exactly a fair deal.
As an adult and mental health nurse, I’ve always known that caring for someone with schizophrenia is about more than just managing their symptoms. Clozapine, often described as the gold standard for treatment-resistant schizophrenia, isn’t without its baggage and it’s a treatment that has been widely covered here in the past.
This new study has a hidden twist: patients on clozapine had a 25% higher risk of infections compared to those on olanzapine, with older adults (55+) being the most vulnerable. Respiratory and gastrointestinal infections stood out as the main culprits.
This study reminds us of the tightrope clinicians walk when prescribing potent medications: balancing psychiatric stability against potential physical health risks.
Clinicians walk a tightrope when prescribing potent medications like clozapine
Methods
This was a population-based cohort study, which used routine data from Hong Kong, covering a period of just over 20 years with people over 18. The primary outcome was the occurrence of any infectious disease, identified by a range of clinical codes, with subtypes analysed as secondary outcomes.
Clozapine was compared against olanzapine because both drugs affect dopamine and serotonin, but clozapine also targets norepinephrine, making it an interesting comparator. The researchers tightened their sample carefully: they excluded patients who hadn’t already tried two antipsychotics (to ensure treatment resistance), those who started both drugs on the same day, anyone who didn’t stick with treatment for at least 90 days, and anyone with a recent history of infection. Routine physical health monitoring was also carried out, in line with practice.
The analysis itself was sophisticated. They used a Cox proportional hazards model to track infection risk over time and adjusted for a wide range of factors, including age, sex, previous medications, coexisting mental health diagnoses, physical health conditions (like diabetes, COPD, and cancer), and other drugs that could influence infection risk. To make the groups more comparable, they applied IPTW (inverse probability of treatment weighting). Results were reported as hazard ratios and absolute rate differences, giving both relative and real-world perspectives. Subgroup analyses broke things down by age and sex, while sensitivity analyses stress-tested the findings by tweaking assumptions, changing treatment duration, using different comparators, or even testing unlikely outcomes like cataracts to check for bias.
Results
This study started with a massive pool of 53,092 people prescribed clozapine or olanzapine in Hong Kong (2004–2023). After applying the inclusion/exclusion rules, the final group included 11,051 patients:
- 1,450 clozapine users
- 9,601 olanzapine users
On average, clozapine users were younger (40.6 vs 45.3 years) and had been exposed to more antipsychotics in the past. The researchers used weighting techniques to balance the two groups, making it a fairer head-to-head comparison.
Main Findings (Infections)
- Total infections recorded: 3,551
- 676 in clozapine users
- 2,875 in olanzapine users
- Incidence rates:
- Clozapine: 7.26 per 1,000 person-years
- Olanzapine: 6.00 per 1,000 person-years
- Risk increase: Clozapine patients had a 25% higher risk of infection (HR = 1.25, 95% CI 1.13 to 1.39).
In plain terms: for every 1,000 patients, clozapine users had about 1.3 more infections each year compared to olanzapine users.
Subgroup Findings
- By gender:
- Men: HR = 1.32 (higher risk)
- Women: HR = 1.18 (still significant, but smaller)
- By age:
- 18 to 44 years: HR = 1.24
- 45 to 54 years: HR = 1.41
- 55+ years: HR = 1.45 (highest risk)
This means infection risk climbs with age: older adults on clozapine are particularly vulnerable.
Clozapine was linked to:
- Upper respiratory infections → HR 1.50
- Lower respiratory infections (like pneumonia) → HR 1.71
- Gastrointestinal infections → HR 1.90
No significant difference was found for sepsis or liver-related infections.
Mortality (1-Year After Infection)
- Upper respiratory infections: very low mortality (≈0.5–0.6%).
- Lower respiratory infections: higher mortality (≈2–3%).
- Gastrointestinal infections: rare deaths, but still important.
Sensitivity Analyses
The researchers really stress-tested their results:
- Tried different timeframes, comparator drugs, and exclusion rules.
- Findings were consistent (HR range: 1.20 to 1.50).
- Even a “negative control” (cataracts) showed no difference, suggesting results weren’t due to random bias.
Clozapine users had 1.3 more infections each year compared to olanzapine users.
Conclusions
Clozapine can be a lifesaver for people with treatment-resistant schizophrenia, but it’s no free pass. This Hong Kong study flags a higher risk of infections, especially in patients over 55, reminding us that physical health can’t take a back seat to mental health (Hu et al., 2025, Fernandez-Egea & McCutcheon 2025).
The authors concluded:
our territory-wide retrospective cohort study in Hong Kong showed an elevated risk of infection among clozapine-treated patients compared with patients using olanzapine, with more pronounced susceptibility to respiratory tract and gastrointestinal infections. The association was stronger in older versus younger patients. Clinicians should balance the therapeutic benefits of clozapine with infection control and other risk-mitigating measures.
This Hong Kong wide study suggests that clinicians should balance the therapeutic benefits of clozapine with infection control and other risk-mitigating measures.
Strengths and limitations
From a critical appraisal point of view, the study ticks many CASP boxes: it used a huge real-world cohort, had clear inclusion criteria, robust analyses, and multiple sensitivity checks. That said, it wasn’t flawless; things like adherence, vaccination status, and lifestyle factors weren’t captured. Even so, the findings remain clinically meaningful.
Strengths of the study
- Large real-world cohort: The study analysed over 10,000 patients, making the findings more generalisable within the Hong Kong population compared to small trials.
- Comparative design: By comparing clozapine to olanzapine (another antipsychotic sometimes used in treatment resistance), the study provided a clinically relevant reference group.
- Robust sensitivity analyses: The use of negative control outcomes, alternative comparators, and age-stratified analyses strengthens the reliability of the results.
- Novel contribution: This is the first population-based cohort to assess clozapine-associated infection risk across a wide range of infectious disease categories.
- Clinical relevance: Findings highlight a specific higher-risk subgroup (older patients) and propose practical risk mitigation strategies such as vaccination and infection surveillance.
- Alignment with existing evidence: Results are consistent with previous cohort studies and the wider literature, adding credibility.
Limitations of the study
- Observational design: This means we cannot establish causality and the results remain vulnerable to bias, even with the strict inclusion criteria.
- Unmeasured confounders: Factors like socioeconomic status, lifestyle, and adherence were not included, which may have influenced results.
- Data gaps: No linked vaccination records, no direct measures of medication compliance, and no lab data on immune cell function or clozapine plasma levels.
- Outcome capture bias: Only infections that required secondary/tertiary care were included, possibly missing milder or primary-care-managed cases.
- Mechanistic uncertainty: Could not disentangle anticholinergic versus immune-related mechanisms due to lack of dosage and co-medication data.
- Population limitations: Conducted in Hong Kong with a predominantly Chinese cohort (slower clozapine metabolism), so findings may not generalise to other ethnicities.
- Lack of patient involvement: No input from people with lived experience, which might have provided valuable context on behavioural or psychosocial pathways.
This is a large, well-conducted real-world cohort study that provides strong evidence for an increased infection risk with clozapine, particularly in older adults. Its strengths lie in sample size, comparative design, and robustness of analyses. However, being observational, it has limitations around causality, unmeasured confounders, data gaps, and generalisability.
This large and well-conducted real-world cohort study provides strong evidence for an increased infection risk with clozapine, but the observational design means we cannot make any conclusions about causation.
Implications for practice
This was a large, carefully adjusted study comparing clozapine and olanzapine in treatment-resistant schizophrenia, focused on infection risk.
The takeaway? Clozapine patients—especially older ones – face higher risks, and this needs to be factored into practice. As an adult and mental health nurse, this continues to demonstrate that holistic care isn’t just a buzzword, it’s essential. We’re not just treating the mind; we’re treating the whole person. That means regular infection screening, vaccination reminders, proactive education, and closer monitoring in older adults.
So, what does this mean for practice?
- We need to keep parity of esteem front and centre: monitoring not just mental health but physical health too.
- Infection prevention should be part of routine care → vaccinations, early screening for cough/fever, education on when to seek help.
- Be especially vigilant with older patients and men on clozapine, who seem to carry the greatest risk.
Clozapine may give someone their life back mentally, but if we ignore their physical vulnerabilities, we’re only doing half the job.
This is parity of esteem in action: you can’t separate physical and mental health, and if we ignore one, we fail the other. Clozapine may improve the brain, but if we aren’t watching the rest of the body, we’re missing half the story.
This study isn’t about scaring people off clozapine. It’s about sharpening our practice. As nurses and clinicians, we need to be upfront with patients about the infection risks, especially those over 55, so they know when to raise the alarm. Prevention is key: keep vaccines up to date, screen regularly, and don’t underestimate the basics like good nutrition and hygiene. Monitoring needs to be smart and individualised. Some patients may need closer follow-up or dose tweaks, particularly if they’ve had repeated infections or are on interacting meds. None of this can be done in isolation: psychiatrists, GPs, pharmacists, nurses, and families all have a role to play in spotting problems early.
And bigger picture implications? Policies and guidelines need to catch up, placing infection monitoring on the same level as the haematological checks we already do. Clozapine can transform lives, but only if we protect the whole person with the same commitment.
Clozapine may give someone their life back mentally, but if we ignore their physical vulnerabilities, we’re only doing half the job.
Links
Primary paper
Yuqi Hu, Wenxin Tian, Cuiling Wei, Qi Sun, Song Song, Lingyue Zhou, Rachel Yui Ki Chu, Wenlong Liu, Boyan Liu, Amy Pui Pui Ng, Krystal Chi Kei Lee, Heidi Ka Ying Lo, Wing Chung Chang, William Chi Wai Wong, Esther Wai Yin Chan, Ian Chi Kei Wong, Francisco Tsz Tsun Lai (2025) Clozapine use and risk of infections in patients with schizophrenia in Hong Kong: a population-based cohort study, The Lancet Psychiatry, Volume 12, Issue 9, Pages 628-637.
Other references
Espejo, G. and Khorassani, F. (2025) ‘Special Report: Clozapine Revisited: Updates to the Gold Standard’, Psychiatric News. 2025/04/29, American Psychiatric Publishing (PN), 60(5).
Fernandez-Egea, E. and McCutcheon, R.A. (2025) ‘Clozapine monitoring requirements: is it time for an update?’, The British Journal of Psychiatry, 226(1), pp. 1–3.
Murray, R. M., Egerton, A., Gao, Y., Grace, A. A., Howes, O., Jauhar, S., Leucht, S., Chen, E. Y. H., MacCabe, J. H., McCutcheon, R. A., Natesan, S., & Taylor, D. (2025). Why Is Clozapine Uniquely Effective in Treatment-Resistant Schizophrenia? A Review and Proposal. Biological Psychiatry, S0006-3223(25)01270-3.
Schneider-Thoma J, Hamza T, Chalkou K, Siafis S, Dong S, Bighelli I, Hansen WP, Scheuring E, Davis JM, Priller J, Baumann P, Conley R, Cordes J, Kelly D, Kluge M, Kumra S, Lewis S, Meltzer HY, Naber D, Schooler N, Volavka J, Wahlbeck K, Salanti G, Leucht S. (2025). Efficacy of clozapine versus second-generation antipsychotics in people with treatment-resistant schizophrenia: a systematic review and individual patient data meta-analysis. Lancet Psychiatry. Apr;12(4):254-265.