Mortality in patients prescribed mirtazapine compared to other antidepressants: an active-comparator new user cohort study

Talk Code: 
Rebecca Joseph
Ruth H Jack, Richard Morriss, Roger David Knaggs, Debbie Butler, Chris Hollis, Julia Hippisley-Cox, Carol Coupland
Author institutions: 
School of Medicine University of Nottingham, National Institute of Health Research Nottingham Biomedical Research Centre Nottingham University Hospitals NHS Trust, School of Pharmacy University of Nottingham, Nuffield Department of Primary Care Health Sciences University of Oxford


There is some evidence for an increased mortality rate in patients prescribed mirtazapine compared to other antidepressants. This study aimed to compare all-cause and cause-specific mortality between patients prescribed mirtazapine and patients prescribed a selective serotonin reuptake inhibitor (SSRI), amitriptyline, or venlafaxine as a second line antidepressant following an initial course of treatment with an SSRI.


The study used English electronic health records provided by the Clinical Practice Research Datalink. Primary care data were linked with mortality and hospital data. The study included patients aged 18-100 years diagnosed with depression who were initially prescribed an SSRI and subsequently prescribed mirtazapine, a different SSRI, amitriptyline, or venlafaxine. The study window was 01 January 2005 – 30 November 2018. The outcomes were all-cause mortality and mortality due to cardiovascular disease, cancer, or respiratory disease (a combined outcome of self-harm and deaths due to suicide is being studied separately). Age-sex standardised mortality rates were calculated. Survival analyses were performed using Cox regression for all-cause mortality and Fine-Gray (competing risks) regression for cause-specific mortality. To meet the proportional hazards assumption, an interaction between treatment group and follow-up time was included in all-cause and cancer mortality models. Differences in baseline characteristics were accounted for using propensity score methods (inverse probability of treatment weighting).


The study included 25,598 patients: 5081 in the mirtazapine group, 15,032 in the SSRI group, 3905 in the amitriptyline group, and 1580 in the venlafaxine group. Median follow-up was 8 months (interquartile range 6.2-18.6) and there were 599 deaths. The crude mortality rate was 16.1 deaths/1000 person-years [95% confidence interval (CI): 14.9-17.4]. The mirtazapine group had the highest standardised mortality rate (21.6 deaths/1000 person-years [95% CI: 18.5-25.0]) and the SSRI group the lowest (13.8 deaths/1000 person-years [95% CI: 12.1-15.6]). In the first two years of follow-up the mirtazapine group had a higher risk of all-cause mortality than the SSRI group (hazard ratio 1.62 [95% CI: 1.28-2.06]), but a similar risk to the amitriptyline (hazard ratio 1.18 [95% CI: 0.85-1.63]) and venlafaxine (hazard ratio 1.11 [95% CI: 0.60-2.05]) groups. A similar pattern was found for deaths due to cancer and respiratory disease. After two years of follow-up , the mirtazapine group had a higher risk of all-cause mortality compared to all three other groups. However, only 20% of patients had over two years of follow-up, and there were few outcomes in this time.


There was an increased risk of mortality in patients prescribed mirtazapine compared to those prescribed an SSRI as second line antidepressant treatments. However, this does not appear strongly driven by a particular cause of death and could reflect residual differences in patient characteristics. Patients prescribed mirtazapine, or other non-SSRI antidepressants, may need support to identify additional health risks and improve their outcomes.

Submitted by: 
Rebecca Joseph
Funding acknowledgement: 
This work has been funded by the National Institute for Health Research (NIHR). The research reported in this paper was conducted by the NIHR Nottingham Biomedical Research Centre. RM is supported by the NIHR MindTech MedTech and in Vitro Fertilisation Collaboration and the NIHR Applied Research Collaboration East Midlands. The views represented are the views of the authors alone and do not necessarily represent the views of the Department of Health in England, the NHS, or the National Institute for Health Research.