Long-term Psychological Effects of Critical Illness

By Saadia R. Akhtar, MD, MSc

Synopsis: This observational study noted that critically ill patients provided with clinical psychological support during their ICU stay had less anxiety, depression, and post traumatic stress disorder at 1 year post-discharge compared to historical controls.

Source: Peris A, Bonizzoli M, Iozzelli D, et al. Early intra-intensive care unit psychological intervention promotes recovery from post traumatic stress disorders, anxiety and depression symptoms in critically ill patients. Crit Care 2011;15:R41.

The authors investigated whether intra-ICU clinical psychological support could impact anxiety, depression, and post-traumatic stress disorder (PTSD) rates in patients 1 year after ICU discharge. This was a single-center study focusing on patients with major trauma (defined by injury severity score > 15) as the primary reason for ICU admission. Inclusion criteria were: age 18-75 years, ICU length of stay (LOS) > 72 hours, mechanical ventilation, ability to be interviewed during ICU stay, and absence of pre-existing psychiatric illness or drug abuse. The study period spanned about 4 years from 2005 to 2009 (the first 2 years for historical control data; the second 2 years for intervention). The psychological interventions provided are described as "educational interventions, counseling, stress management [cognitive and emotional restructuring], psychological support, coping strategies designed to ease the management of anxiety, depression, fear, hopelessness ..." These were provided several times a day by clinical psychologists who were available in-house from 12 a.m. to 4 p.m. and also by other ICU staff. Similar but separate interventions were administered to patients' family members. Validated standard questionnaires were used for assessment of PTSD (Impact of Event Scale Revised, IESR), anxiety and depression (Hospital Anxiety and Depression Scale, HADS), and quality of life.

In the 4-year study period, 376 patients met inclusion criteria and, of these, based on availability and willingness for interview and follow-up, 86 were enrolled in the control arm and 123 in the intervention arm. The two groups were well matched in demographic and diagnostic features as well as ICU course and LOS. At 1 year after ICU discharge, patients in the intervention group were less likely to have anxiety or depression by HADS score, but results did not reach statistical significance. They were significantly less likely to have PTSD (21% vs. 57%) or need anxiolytics or antidepressants; they also had better overall subjective assessment of quality of life. Logistic regression with some predefined variables and some selected post hoc variables found no clear predictors of long-term anxiety, depression, or PTSD; small associations with Glasgow coma scale at admission and ICU discharge were noted.

Psychological effects of critical illness and ICU care are an extremely important, but poorly understood and studied aspect of intensive care medicine. Rates of PTSD are high though variable, estimated at 20% in one meta-analysis of 15 studies of general ICU populations, 28% in survivors of acute lung injury and upwards of 60% in other reports.1 Symptoms such as long-term anxiety and depression are similarly commonly noted. Thus Peris et al are to be commended for considering this issue and trying to provide an intervention that may improve psychological outcomes after critical illness; they are the first investigators to do this.

The study has several limitations. One key deficiency is that the interventions provided are not clearly defined or documented in the report, either in terms of the methods or the time spent per patient and family member; this will make it difficult for others to repeat the study or apply the interventions. There are several issues with the study design that limit the accuracy, validity, and utility of the results; use of historical controls, lack of a priori planning of sample size, thus inadequate powering for assessment of outcomes and predictive factors for anxiety, depression, and PTSD, and absence of specific data about sedative and analgesic use are some examples. There are considerable differences in the rates of PTSD observed in the historical controls here compared to rates reported for trauma ICU populations in other publications; as a result, the observed treatment effect may be exaggerated.

Despite these issues, this remains an important and at least hypothesis-generating pioneer study; it suggests that there may be some positive outcomes from early psychological support/intervention for patients and families in the ICU. It also reminds us that considerable additional work is needed to understand the factors that predispose or contribute to the development of anxiety, depression, and PTSD in critically ill patients; targeted interventions based on such data may be most effective. I can only hope that there will soon be several more robustly designed investigations into this topic that will provide clear answers and direction to guide preventive care. In the meantime, I suggest we continue to acknowledge the short- and long-term psychological side effects of critical care and provide as much general support as possible to our patients and their families.


1. Davydow DS, Gifford JM, Desai SV, et al. Posttraumatic stress disorder in general intensive care unit survivors: A systematic review. Gen Hosp Psychiatry 2008;30:421-434.