ICU (Intensive Care Unit) admission rates are high and are expected to continue to increase. Critical care advances have improved the chance of survival in recent years. Understanding the experiences and needs of ICU survivors is essential to improving their care.
The physical health (e,g., fatigue, pain, cognitive impairment, muscle weakness; Azoulay et al., 2017), mental health (e.g., post-traumatic stress, anxiety, psychosis, depression, and substance use; Prince et al., 2018; Sareen et al., 2020) and overall quality of life (Oeyen et al., 2010) is often impacted in survivors of critical illness. Such a combination of difficulties may feel understandably overwhelming and lead ICU survivors to engage in self-harm and contemplate and/or attempt suicide. However, little is known about the prevalence of self-harm and death by suicide in this population. As such, Fernando et el. (2021) aimed to address this gap by investigating the relationship between survival and self-harm or suicide following ICU admission.
The authors used a large population-level cohort study design. Data regarding the cause of death, demographics, admission and discharge dates, and a number of hospital admissions were extracted from Ontario (Canada) healthcare administrative databases. The study did not require ethical approval as data were collected through ICES who are permitted to collect and analyse such data under section 45 of the Personal Health Information Protection Act of Ontario.
Participants were included if they were adults, survived ICU and/or general hospital admission and discharge. ICD-10 codes were used to identify the frequency of death by suicide and the ICES Mental Health and Addictions Scorecard and Evaluation Framework was used to identify hospital visits for self-harm. The primary outcome was composite rates for death by suicide or self-harm requiring a hospital visit. These factors were also analysed separately.
Data were analysed using descriptive statistics to measure central tendency (means and medians) and variability (standard deviations and interquartile ranges). Cumulative incidence function curves were used to collate instances of suicide and self-harm. Comparisons in death by suicide and self-harm were made between ICU and non-ICU hospital survivors using overlap propensity score weighting.
The sample included 423,060 ICU survivors and 3,081,111 non-ICU hospital survivors identified between 1 January 2009 and 31 December 2017. The majority of ICU survivors were older and fewer were female compared to non-ICU hospital survivors. There was a high prevalence of co-morbid health conditions and mental health problems in ICU survivors, with a high proportion receiving invasive mechanical ventilation.
Comparison between ICU survivors and non-ICU survivors:
- 750 ICU survivors (0.2%) died by suicide, whereas 2,427 non-ICU hospital survivors (0.1%) died by suicide
- 5,662 ICU survivors (1.3%) had self-harmed after discharge, compared to 24,411 non-ICU hospital survivors (0.8%)
- Collectively, 6,234 ICU survivors (1.5%) died by suicide or self-harmed after discharge, in contrast to 26,376 non-ICU hospital survivors (0.9%)
- Overall, ICU admission was related to a higher risk of death by suicide, self-harm, or a combination of both, than non-ICU hospital admission. This increased risk was observed quite rapidly after discharge and persisted over the years.
Comparison between ICU survivors who died by suicide or self-harmed and those who did not:
- ICU survivors who died by suicide or self-harmed post-discharge were:
- lived in socioeconomically disadvantaged areas,
- had fewer co-morbidities,
- experienced mental health problems prior to admission,
- were given invasive mechanical ventilation or renal replacement therapy, and
- discharged directly home
- Lower risk of suicide was related to:
- older age,
- living in more affluent areas, and
- being discharged to a rehabilitation or long-term care setting.
Overall, this article highlighted the increased risk of suicide and self-harm for ICU survivors post-discharge. Several factors were identified to further exacerbate the risk of suicide and self-harm in this population.
Strengths and limitations
A particular strength of the study is the large population cohort design. The entire population was included and there was little missing data, which adds to the trustworthiness of the study and indicates that these findings are applicable to Ontario, Canada. However, it may be difficult to apply these findings to other areas of Canada, countries and cultures. Furthermore, the study omitted ethnicity, sexuality, and gender information. This further limits the relevance of findings and it is not clear how these additional intersections in identity might influence suicide and self-harm rates in ICU survivors.
The authors acknowledged the lack of patient and public involvement in the study to protect confidentiality. However, there is a question as to whether representatives with lived experience of ICU admission could have contributed to the study design, interpretation of the data and manuscript creation by signing a non-disclosure confidentiality agreement alongside the study team. It is important to involve those with lived experience to improve the quality and relevance of the research (INVOLVE, 2021). Therefore, future studies should involve those with lived experience, especially if investigating possible early intervention strategies.
Implications for practice
Having had some experience working in a ventilation service during a health placement on the doctorate in clinical psychology, I have seen the benefits and opportunities for patients to be routinely screened by members of the MDT (multi-disciplinary team) and offered psychology on transfer from ICU to the ventilation unit for tracheostomy (invasive ventilation) weaning.
The findings of the current study indicate particular characteristics that staff could identify and flag within ward round and MDT when considering discharge planning. Intervening early may help to reduce the risk of suicide and self-harm in ICU survivors post-discharge. Follow-up outpatient interventions, more specifically, physiotherapy, have shown the potential to improve low mood and quality of life in the short term. Additionally, psychological and medical interventions have reduced post-traumatic stress symptoms in the medium term post-discharge (Rosa et al., 2019). However, suicidal experiences were not measured and so the impact on suicidal experiences is unclear. Therefore, future studies should focus on developing and evaluating early intervention strategies to reduce suicide and self-harm in ICU survivors post-discharge.
Statement of interests
Fernando, S. M., Qureshi, D., Sood, M. M., Pugliese, M., Talarico, R., Myran, D. T., … & Kyeremanteng, K. (2021). Suicide and self-harm in adult survivors of critical illness: population based cohort study. BMJ (Clinical research ed.), 373, n973. https://doi.org/10.1136/bmj.n973
Azoulay, E., Vincent, J. L., Angus, D. C., Arabi, Y. M., Brochard, L., Brett, S. J., … Herridge, M. (2017). Recovery after critical illness: putting the puzzle together-a consensus of 29. Critical care (London, England), 21(1), 296. https://doi.org/10.1186/s13054-017-1887-7
INVOLVE (2021). Briefing notes for researchers – public involvement in NHS, health and social care research. The NIHR website, last accessed 26 Mar 2022.
Oeyen, S. G., Vandijck, D. M., Benoit, D. D., Annemans, L., & Decruyenaere, J. M. (2010). Quality of life after intensive care: a systematic review of the literature. Critical care medicine, 38(12), 2386–2400. https://doi.org/10.1097/CCM.0b013e3181f3dec5
Prince, E., Gerstenblith, T. A., Davydow, D., & Bienvenu, O. J. (2018). Psychiatric Morbidity After Critical Illness. Critical care clinics, 34(4), 599–608. https://doi.org/10.1016/j.ccc.2018.06.006
Rosa, R. G., Ferreira, G. E., Viola, T. W., Robinson, C. C., Kochhann, R., Berto, P. P., … & Teixeira, C. (2019). Effects of post-ICU follow-up on subject outcomes: A systematic review and meta-analysis. Journal of critical care, 52, 115–125. https://doi.org/10.1016/j.jcrc.2019.04.014
Sareen, J., Olafson, K., Kredentser, M. S., Bienvenu, O. J., Blouw, M., Bolton, J. M., … & Marrie, R. A. (2020). The 5-Year Incidence of Mental Disorders in a Population-Based ICU Survivor Cohort. Critical care medicine, 48(8), e675–e683. https://doi.org/10.1097/CCM.0000000000004413